View source for Engineering:Artificial intelligence

{{Short description|Ability of systems to perceive, synthesize, and infer information}}
{{Artificial intelligence}}

<!-- DEFINITIONS -->
'''Artificial intelligence''' ('''AI''') is [[Philosophy:Intelligence|intelligence]] demonstrated by machines, as opposed to [[Philosophy:Human intelligence|intelligence displayed by humans]] or [[Biology:Animal cognition|by other animals]]. "Intelligence" encompasses the ability to learn and to reason, to generalize, and to infer meaning.<ref>{{cite web |title=Artificial intelligence (AI) |url=https://www.britannica.com/technology/artificial-intelligence |website=www.britannica.com |access-date=18 June 2023 |language=en |date=16 June 2023 |quote= The term is frequently applied to the project of developing systems endowed with the intellectual processes characteristic of humans, such as the ability to reason, discover meaning, generalize, or learn from past experience.}}</ref> Example tasks in which this is done include speech recognition, computer vision, translation between (natural) languages, as well as other mappings of inputs.<ref>{{cite book |last1=Winston |first1=P H |title=Artificial intelligence. Second edition |date=1984 |location=United States |url=https://dl.acm.org/doi/abs/10.5555/30424}}</ref>

<!-- SUMMARIZING APPLICATIONS -->
AI applications include advanced [[Web search|web search]] engines (e.g., [[Software:Google Search|Google Search]]), [[Recommender system|recommendation systems]] (used by [[Company:YouTube|YouTube]], [[Company:Amazon|Amazon]], and [[Company:Netflix|Netflix]]), [[Natural-language understanding|understanding human speech]] (such as [[Software:Siri|Siri]] and [[Software:Amazon Alexa|Alexa]]), [[Engineering:Self-driving car|self-driving car]]s (e.g., [[Company:Waymo|Waymo]]), [[Generative artificial intelligence|generative]] or [[Computational creativity|creative]] tools ([[Software:ChatGPT|ChatGPT]] and AI art), [[Automated decision-making|automated decision-making]], and competing at the highest level in strategic game systems (such as chess and [[Software:Go (game)|Go]]).{{sfnp|Google|2016}}

As machines become increasingly capable, tasks considered to require "intelligence" are often removed from the definition of AI, a phenomenon known as the [[Philosophy:AI effect|AI effect]].{{sfnp|McCorduck|2004|p=204}} For instance, [[Optical character recognition|optical character recognition]] is frequently excluded from things considered to be AI, having become a routine technology.{{sfnp|Schank|1991|page=38}}

<!-- SUMMARIZING HISTORY AND APPROACHES  -->
Artificial intelligence was founded as an academic discipline in 1956, and in the years since it has experienced several waves of optimism,{{sfnp|Crevier|1993|p=109}}<ref name="AI in the 80s" /> followed by disappointment and the loss of funding (known as an "[[AI winter]]"),<ref name="First AI winter" /><ref name="Second AI winter" /> followed by new approaches, success, and renewed funding.<ref name="AI in the 80s" />{{sfnp|Clark|2015b}} AI research has tried and discarded many different approaches, including simulating the brain, modeling human problem solving, formal logic, large databases of knowledge, and imitating animal behavior. In the first decades of the 21st century, highly mathematical and statistical [[Machine learning|machine learning]] has dominated the field, and this technique has proved highly successful, helping to solve many challenging problems throughout industry and academia.{{sfnp|Clark|2015b}}<ref name="AI widely used 1990s" />

<!-- SUMMARIZING GOALS AND TOOLS: SCOPE OF AI  -->
The various sub-fields of AI research are centered around particular goals and the use of particular tools. The traditional goals of AI research include [[Automated reasoning|reasoning]], knowledge representation, [[Automated planning and scheduling|planning]], [[Machine learning|learning]], [[Natural language processing|natural language processing]], [[Machine perception|perception]], and the ability to move and manipulate objects.{{efn|name="Problems of AI"}} [[Artificial general intelligence|General intelligence]] (the ability to solve an arbitrary problem) is among the field's long-term goals.<ref name="Artificial General Intelligence" /> To solve these problems, AI researchers have adapted and integrated a wide range of problem-solving techniques, including search and mathematical optimization, formal logic, [[Artificial neural network|artificial neural network]]s, and methods based on [[Statistics|statistics]], [[Probability|probability]], and [[Finance:Economics|economics]]. AI also draws upon [[Computer science|computer science]], [[Social:Psychology|psychology]], [[Social:Linguistics|linguistics]], [[Philosophy:Philosophy|philosophy]], and many other fields.

<!-- SUMMARISING PHILOSOPHY AND FUTUROLOGY -->
The field was founded on the assumption that human intelligence "can be so precisely described that a machine can be made to simulate it".{{efn|This statement comes from the proposal for the [[Dartmouth workshop]] of 1956, which reads: "Every aspect of learning or any other feature of intelligence can be so precisely described that a machine can be made to simulate it."{{sfnp|McCarthy|Minsky|Rochester|Shannon|1955}}}} This raised philosophical arguments about the mind and the ethical consequences of creating artificial beings endowed with human-like intelligence; these issues have previously been explored by [[History of artificial intelligence#Precursors|myth]], fiction (science fiction), and [[Philosophy:Philosophy of artificial intelligence|philosophy]] since antiquity.{{sfnp|Newquist|1994|pp=45–53}} [[Computer scientist]]s and [[Biography:Philosopher|philosopher]]s have since suggested that AI may become an existential risk to humanity if its rational capacities are not steered towards goals beneficial to humankind.{{Efn|Russel and Norvig note in the textbook Artificial Intelligence: A Modern Approach (4th ed.), section 1.5:
"In the longer term, we face the difficult problem of controlling superintelligent AI systems that may evolve in unpredictable ways." while referring to computer scientists, philosophers, and technologists.|name="Future risks of AI"}} Economists have frequently highlighted the risks of redundancies from AI, and speculated about unemployment if there is no adequate social policy for full employment.<ref>E McGaughey, 'Will Robots Automate Your Job Away? Full Employment, Basic Income, and Economic Democracy' (2022) [https://academic.oup.com/ilj/article/51/3/511/6321008 51(3) Industrial Law Journal 511–559]</ref> The term artificial intelligence has also been criticized for overhyping AI's true technological capabilities.<ref>{{cite news|url=https://www.analyticsinsight.net/is-ai-overhyped-in-2022-getting-the-truth-about-the-true-power/|title=s AI Overhyped in 2022? Getting the Truth About the True Power|publisher=Analytics Insight|date=21 March 2022|accessdate=11 March 2023|archive-date=10 March 2023|archive-url=https://web.archive.org/web/20230310113530/https://www.analyticsinsight.net/is-ai-overhyped-in-2022-getting-the-truth-about-the-true-power/|url-status=live}}</ref><ref>{{cite news|url=https://www.technologyreview.com/2018/09/13/240156/artificial-intelligence-is-often-overhypedand-heres-why-thats-dangerous/|title=Artificial intelligence is often overhyped—and here's why that's dangerous|first=Martin|last=Giles|publisher=MIT Technology|date=13 September 2018|accessdate=11 March 2023|archive-date=11 March 2023|archive-url=https://web.archive.org/web/20230311145742/https://www.technologyreview.com/2018/09/13/240156/artificial-intelligence-is-often-overhypedand-heres-why-thats-dangerous/|url-status=live}}</ref><ref>{{cite news|url=https://www.cbc.ca/radio/sunday/the-sunday-edition-for-february-23-2020-1.5468283/is-ai-overhyped-researchers-weigh-in-on-technology-s-promise-and-problems-1.5468289|title=Is AI overhyped? Researchers weigh in on technology's promise and problems|first=Ira|last=Basen|publisher=Canadian Broadcasting Corporation|date=21 February 2020|accessdate=11 March 2023|archive-date=11 March 2023|archive-url=https://web.archive.org/web/20230311145743/https://www.cbc.ca/radio/sunday/the-sunday-edition-for-february-23-2020-1.5468283/is-ai-overhyped-researchers-weigh-in-on-technology-s-promise-and-problems-1.5468289|url-status=live}}</ref>

== History ==
{{Main|History of artificial intelligence|Timeline of artificial intelligence}}

[[File:Didrachm Phaistos obverse CdM.jpg|thumb|Silver didrachma from Crete depicting Talos, a mythical intelligent automaton ({{c.}} 300 BC)]]

<!-- PRE-20TH CENTURY. MAYBE TO BE KEPT SHORT. -->
Artificial beings with intelligence appeared as storytelling devices in antiquity,<ref name="AI in myth">
AI in myth:
* {{Harvtxt|McCorduck|2004|pp=4–5}}
* {{Harvtxt|Russell|Norvig|2003|p=939}}
</ref> and have been common in fiction, as in Mary Shelley's ''Frankenstein'' or Karel Čapek's ''R.U.R.''<!-- PLEASE DON'T ADD MORE EXAMPLES. THIS IS ENOUGH. SEE SECTION AT BOTTOM OF ARTICLE ON FICTION.-->{{sfnp|McCorduck|2004|pp=17–25}} These characters and their fates raised many of the same issues now discussed in the [[Philosophy:Ethics of artificial intelligence|ethics of artificial intelligence]].{{sfnp|McCorduck|2004|pp=340–400}}

<!-- MAJOR INTELLECTUAL PRECURSORS: LOGIC, THEORY OF COMPUTATION, CYBERNETICS, INFORMATION, NEUROBIOLOGY: Antiquity - 1955 -->
The study of mechanical or "formal" reasoning began with [[Biography:Philosopher|philosopher]]s and mathematicians in antiquity. The study of mathematical logic led directly to [[Biography:Alan Turing|Alan Turing]]'s [[Theory of computation|theory of computation]], which suggested that a machine, by shuffling symbols as simple as "0" and "1", could simulate any conceivable act of mathematical deduction. This insight that digital computers can simulate any process of formal reasoning is known as the [[Church–Turing thesis]].{{sfnp|Berlinski|2000}} This, along with concurrent discoveries in [[Biology:Neuroscience|neurobiology]], [[Information theory|information theory]] and [[Cybernetics|cybernetics]], led researchers to consider the possibility of building an electronic brain.<ref>
AI's immediate precursors:
* {{Harvtxt|McCorduck|2004|pp=51–107}}
* {{Harvtxt |Crevier|1993|pp=27–32}}
* {{Harvtxt |Russell|Norvig|2003|pp=15, 940}}
* {{Harvtxt |Moravec|1988|p=3}}
</ref> The first work that is now generally recognized as AI was McCullouch and [[Biography:Walter Pitts|Pitts]]' 1943 formal design for Turing-complete "artificial neurons".{{sfnp|Russell|Norvig|2009|p=16}}

<!-- SYMBOLIC AI APPROACH: 1955-1990 -->
By the 1950s, two visions for how to achieve machine intelligence emerged. One vision, known as Symbolic AI or [[GOFAI]], was to use computers to create a symbolic representation of the world and systems that could reason about the world. Proponents included [[Biography:Allen Newell|Allen Newell]], [[Biography:Herbert A. Simon|Herbert A. Simon]], and [[Biography:Marvin Minsky|Marvin Minsky]]. Closely associated with this approach was the [[Heuristic (computer science)|"heuristic search"]] approach, which likened intelligence to a problem of exploring a space of possibilities for answers.

The second vision, known as the [[Connectionism|connectionist approach]], sought to achieve intelligence through learning. Proponents of this approach, most prominently [[Biography:Frank Rosenblatt|Frank Rosenblatt]], sought to connect Perceptron in ways inspired by connections of neurons.{{sfn|Manyika|2022|p=9}} James Manyika and others have compared the two approaches to the mind (Symbolic AI) and the brain (connectionist). Manyika argues that symbolic approaches dominated the push for artificial intelligence in this period, due in part to its connection to intellectual traditions of Descartes, Boole, [[Biography:Gottlob Frege|Gottlob Frege]], [[Biography:Bertrand Russell|Bertrand Russell]], and others. Connectionist approaches based on [[Cybernetics|cybernetics]] or [[Artificial neural network|artificial neural network]]s were pushed to the background but have gained new prominence in recent decades.{{sfn|Manyika|2022|p=10}}

<!-- 1956 - 1974 -->
The field of AI research was born at [[Dartmouth workshop|a workshop]] at [[Organization:Dartmouth College|Dartmouth College]] in 1956.{{efn|
Daniel Crevier wrote, "the conference is generally recognized as the official birthdate of the new science."{{sfnp|Crevier|1993|pp=47–49}} [[Biography:Stuart J. Russell|Russell]] and [[Biography:Peter Norvig|Norvig]] call the conference "the birth of artificial intelligence."{{sfnp|Russell|Norvig|2003|p=17}}}}<ref>
[[Dartmouth workshop]]:
* {{Harvtxt |Russell|Norvig|2003|p=17}}
* {{Harvtxt|McCorduck|2004|pp=111–136}}
* {{Harvtxt|NRC|1999|pp=200–201}}
The proposal:
* {{Harvtxt|McCarthy|Minsky|Rochester|Shannon|1955}}
</ref> The attendees became the founders and leaders of AI research.{{efn|
[[Biography:Stuart J. Russell|Russell]] and [[Biography:Peter Norvig|Norvig]] wrote "for the next 20 years the field would be dominated by these people and their students."{{sfnp|Russell|Norvig|2003|p=17}}
}} They and their students produced programs that the press described as "astonishing":{{efn|
[[Biography:Stuart J. Russell|Russell]] and [[Biography:Peter Norvig|Norvig]] wrote "it was astonishing whenever a computer did anything kind of smartish".{{sfnp|Russell|Norvig|2003|p=18}}
}} computers were learning checkers strategies, solving word problems in algebra, proving [[Theorem|logical theorems]] and speaking English.{{efn|
The programs described are [[Biography:Arthur Samuel (computer scientist)|Arthur Samuel]]'s checkers program for the IBM 701, Daniel Bobrow's [[Software:STUDENT (computer program)|STUDENT]], [[Biography:Allen Newell|Newell]] and [[Biography:Herbert A. Simon|Simon]]'s [[Logic Theorist]] and [[Biography:Terry Winograd|Terry Winograd]]'s [[Software:SHRDLU|SHRDLU]].
}}<ref>
Successful Symbolic AI programs:
* {{Harvtxt|McCorduck|2004|pp=243–252}}
* {{Harvtxt|Crevier|1993|pp=52–107}}
* {{Harvtxt|Moravec|1988|p=9}}
* {{Harvtxt|Russell|Norvig|2003|pp=18–21}}
</ref>

By the middle of the 1960s, research in the U.S. was heavily funded by the Department of Defense<ref>
AI heavily funded in the 1960s:
* {{Harvtxt|McCorduck|2004|p=131}}
* {{Harvtxt|Crevier|1993|pp=51, 64–65}}
* {{Harvtxt|NRC|1999|pp=204–205}}
</ref> and laboratories had been established around the world.{{sfnp|Howe|1994}}

<!-- 1965-75 Excessive Optimism to AI Winter -->
Researchers in the 1960s and the 1970s were convinced that symbolic approaches would eventually succeed in creating a machine with [[Artificial general intelligence|artificial general intelligence]] and considered this the goal of their field.{{sfnp|Newquist|1994|pp=86–86}} [[Biography:Herbert A. Simon|Herbert Simon]] predicted, "machines will be capable, within twenty years, of doing any work a man can do".<ref>
{{Harvtxt|Simon|1965|p=96}} quoted in {{Harvtxt|Crevier|1993|p=109}}
</ref> [[Biography:Marvin Minsky|Marvin Minsky]] agreed, writing, "within a generation&nbsp;... the problem of creating 'artificial intelligence' will substantially be solved".<ref>
{{Harvtxt|Minsky|1967|p=2}} quoted in {{Harvtxt|Crevier|1993|p=109}}
</ref>

They had failed to recognize the difficulty of some of the remaining tasks. Progress slowed and in 1974, in response to the [[Lighthill report|criticism]] of Sir James Lighthill{{sfnp|Lighthill|1973}} and ongoing pressure from the US Congress to fund more productive projects, both the U.S. and British governments cut off exploratory research in AI. The next few years would later be called an "[[AI winter]]", a period when obtaining funding for AI projects was difficult.<ref name="First AI winter">
First AI Winter, [[Lighthill report]], Mansfield Amendment
* {{Harvtxt|Crevier|1993|pp=115–117}}
* {{Harvtxt|Russell|Norvig|2003|p=22}}
* {{Harvtxt|NRC|1999|pp=212–213}}
* {{Harvtxt|Howe|1994}}
* {{Harvtxt|Newquist|1994|pp=189–201}}
</ref>

<!-- 1980s -->
In the early 1980s, AI research was revived by the commercial success of [[Expert system|expert system]]s,<ref>
Expert systems:
* {{Harvtxt|Russell|Norvig|2003|pp=22–24}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=227–331}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 17.4}}
* {{Harvtxt|McCorduck|2004|pp=327–335, 434–435}}
* {{Harvtxt|Crevier|1993|pp=145–62, 197–203}}
* {{Harvtxt|Newquist|1994|pp=155–183}}
</ref> a form of AI program that simulated the knowledge and analytical skills of human experts. By 1985, the market for AI had reached over a billion dollars. At the same time, Japan's [[Fifth generation computer|fifth generation computer]] project inspired the U.S. and British governments to restore funding for academic research.<ref name="AI in the 80s">
Funding initiatives in the early 80s: Fifth Generation Project (Japan), [[Alvey]] (UK), [[Company:Microelectronics and Computer Technology Corporation|Microelectronics and Computer Technology Corporation]] (US), [[Strategic Computing Initiative]] (US):
* {{Harvtxt|McCorduck|2004|pp=426–441}}
* {{Harvtxt|Crevier|1993|pp=161–162,197–203, 211, 240}}
* {{Harvtxt|Russell|Norvig|2003|p=24}}
* {{Harvtxt|NRC|1999|pp=210–211}}
* {{Harvtxt|Newquist|1994|pp=235–248}}
</ref> However, beginning with the collapse of the Lisp Machine market in 1987, AI once again fell into disrepute, and a second, longer-lasting winter began.<ref name="Second AI winter">
Second AI Winter:
* {{Harvtxt|McCorduck|2004|pp=430–435}}
* {{Harvtxt|Crevier|1993|pp=209–210}}
* {{Harvtxt|NRC|1999|pp=214–216}}
* {{Harvtxt|Newquist|1994|pp=301–318}}
</ref>

<!-- SUBSYMBOLIC APPROACHES, the 1980s -->
Many researchers began to doubt that the symbolic approach would be able to imitate all the processes of human cognition, especially [[Machine perception|perception]], robotics, [[Machine learning|learning]] and [[Pattern recognition|pattern recognition]]. A number of researchers began to look into "sub-symbolic" approaches to specific AI problems.{{sfnp|Nilsson|1998|p=7}} [[Engineering:Robotics|Robotics]] researchers, such as [[Biography:Rodney Brooks|Rodney Brooks]], rejected symbolic AI and focused on the basic engineering problems that would allow robots to move, survive, and learn their environment.{{efn|
Embodied approaches to AI{{sfnp|McCorduck|2004|pp=454–462}} were championed by [[Biography:Hans Moravec|Hans Moravec]]{{sfnp|Moravec|1988}} and [[Biography:Rodney Brooks|Rodney Brooks]]{{sfnp|Brooks|1990}} and went by many names: [[Nouvelle AI]],{{sfnp|Brooks|1990}} [[Developmental robotics]],<ref name = "Developmental robotics">
[[Developmental robotics]]:
* {{Harvtxt|Weng|McClelland|Pentland|Sporns|2001}}
* {{Harvtxt|Lungarella|Metta|Pfeifer|Sandini|2003}}
* {{Harvtxt|Asada|Hosoda|Kuniyoshi|Ishiguro|2009}}
* {{Harvtxt|Oudeyer|2010}}
</ref>
[[Situated|situated]] AI, behavior-based AI as well as others. A similar movement in cognitive science was the embodied mind thesis.
}}

Interest in [[Artificial neural network|neural networks]] and "[[Connectionism|connectionism]]" was revived by [[Biography:Geoffrey Hinton|Geoffrey Hinton]], [[Biography:David Rumelhart|David Rumelhart]] and others in the middle of the 1980s.<ref>
Revival of connectionism:
* {{Harvtxt|Crevier|1993|pp=214–215}}
* {{Harvtxt|Russell|Norvig|2003|p=25}}
</ref> [[Soft computing]] tools were developed in the 1980s, such as [[Artificial neural network|neural networks]], fuzzy systems, Grey system theory, [[Evolutionary computation|evolutionary computation]] and many tools drawn from [[Statistics|statistics]] or [[Mathematical optimization|mathematical optimization]].

<!-- FORMAL METHODS RISING IN THE 1990s: "Statistical AI"  -->
AI gradually restored its reputation in the late 1990s and early 21st century by finding specific solutions to specific problems. The narrow focus allowed researchers to produce verifiable results, exploit more mathematical methods, and collaborate with other fields (such as [[Statistics|statistics]], [[Finance:Economics|economics]] and [[Mathematical optimization|mathematics]]).<ref name="AI 1990s">
Formal and narrow methods adopted in the 1990s:
* {{Harvtxt |Russell|Norvig|2003|pp=25–26}}
* {{Harvtxt|McCorduck|2004|pp=486–487}}
</ref> By 2000, solutions developed by AI researchers were being widely used, although in the 1990s they were rarely described as "artificial intelligence".<ref name="AI widely used 1990s">
AI widely used in the late 1990s:
* {{Harvtxt|Russell|Norvig|2003|p=28}}
* {{Harvtxt|Kurzweil|2005|p=265}}
* {{Harvtxt|NRC|1999|pp=216–222}}
* {{Harvtxt|Newquist|1994|pp=189–201}}
</ref>

<!--DEEP LEARNING, BIG DATA & MACHINE LEARNING IN THE 2010s-present -->
[[Moore's law|Faster computers]], algorithmic improvements and access to [[Big data|large amounts of data]] enabled advances in [[Machine learning|machine learning]] and perception; data-hungry [[Deep learning|deep learning]] methods started to dominate accuracy benchmarks [[Deep learning#Deep learning revolution|around 2012]].{{sfnp|McKinsey|2018}} According to {{wipe|Bloomberg News}} Jack Clark, 2015 was a landmark year for artificial intelligence, with the number of software projects that use AI within [[Company:Google|Google]] increased from a "sporadic usage" in 2012 to more than 2,700 projects.{{efn|
Clark wrote: "After a half-decade of quiet breakthroughs in artificial intelligence, 2015 has been a landmark year. Computers are smarter and learning faster than ever."{{sfnp|Clark|2015b}}
}} He attributed this to an increase in affordable [[Artificial neural network|neural networks]], due to a rise in cloud computing infrastructure and to an increase in research tools and datasets.{{sfnp|Clark|2015b}}

In a 2017 survey, one in five companies reported they had "incorporated AI in some offerings or processes".<ref>{{Harvtxt|MIT Sloan Management Review|2018}}; {{Harvtxt|Lorica|2017}}</ref> The amount of research into AI (measured by total publications) increased by 50% in the years 2015–2019.{{sfnp|UNESCO|2021}} According to AI Impacts at Stanford, around 2022 about $50 billion annually is invested in artificial intelligence in the US, and about 20% of new US Computer Science PhD graduates have specialized in artificial intelligence;<ref>{{cite news |last1=DiFeliciantonio |first1=Chase |title=AI has already changed the world. This report shows how |url=https://www.sfchronicle.com/tech/article/ai-artificial-intelligence-report-stanford-17869558.php |access-date=19 June 2023 |work=San Francisco Chronicle |date=3 April 2023}}</ref> about 800,000 AI-related US job openings existed in 2022.<ref>{{cite news |last1=Goswami |first1=Rohan |title=Here's where the A.I. jobs are |url=https://www.cnbc.com/2023/04/05/ai-jobs-see-the-state-by-state-data-from-a-stanford-study.html |access-date=19 June 2023 |work=CNBC |date=5 April 2023 |language=en}}</ref>

<!-- AGI, 2005-present -->
Numerous academic researchers became concerned that AI was no longer pursuing the original goal of creating versatile, fully intelligent machines. Much of current research involves statistical AI, which is overwhelmingly used to solve specific problems, even highly successful techniques such as [[Deep learning|deep learning]]. This concern has led to the subfield of [[Artificial general intelligence|artificial general intelligence]] (or "AGI"), which had several well-funded institutions by the 2010s.<ref name="Artificial General Intelligence">
{{Harvtxt|Pennachin|Goertzel|2007}}; {{Harvtxt|Roberts|2016}}
</ref>

== Goals ==
The general problem of simulating (or creating) intelligence has been broken down into sub-problems. These consist of particular traits or capabilities that researchers expect an intelligent system to display. The traits described below have received the most attention.{{efn|name="Problems of AI"|This list of intelligent traits is based on the topics covered by the major AI textbooks, including: {{Harvtxt|Russell|Norvig|2003}}, {{Harvtxt|Luger|Stubblefield|2004}}, {{Harvtxt|Poole|Mackworth|Goebel|1998}} and {{Harvtxt|Nilsson|1998}}}}

=== Reasoning, problem-solving ===
<!-- This is linked to in the introduction --><!-- SOLVED PROBLEMS -->
Early researchers developed algorithms that imitated step-by-step reasoning that humans use when they solve puzzles or make logical deductions.<ref>
Problem solving, puzzle solving, game playing and deduction:
* {{Harvtxt|Russell|Norvig|2003|loc=chpt. 3–9}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|loc=chpt. 2,3,7,9}}
* {{Harvtxt|Luger|Stubblefield|2004|loc=chpt. 3,4,6,8}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 7–12}}
</ref> By the late 1980s and 1990s, AI research had developed methods for dealing with [[Physics:Uncertainty|uncertain]] or incomplete information, employing concepts from [[Probability|probability]] and [[Finance:Economics|economics]].<ref>
Uncertain reasoning:
* {{Harvtxt|Russell|Norvig|2003|pp=452–644}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=345–395}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=333–381}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 19}}
</ref>

Many of these algorithms proved to be insufficient for solving large reasoning problems because they experienced a "combinatorial explosion": they became exponentially slower as the problems grew larger.<ref name="Intractability">
Intractability and efficiency and the [[Combinatorial explosion|combinatorial explosion]]:
* {{Harvtxt|Russell|Norvig|2003|pp=9, 21–22}}
</ref> Even humans rarely use the step-by-step deduction that early AI research could model. They solve most of their problems using fast, intuitive judgments.<ref name="Psychological evidence of sub-symbolic reasoning">
Psychological evidence of the prevalence sub-symbolic reasoning and knowledge:
* {{Harvtxt|Kahneman|2011}}
* {{Harvtxt|Wason|Shapiro|1966}}
* {{Harvtxt|Kahneman|Slovic|Tversky|1982}}
* {{Harvtxt|Dreyfus|Dreyfus|1986}}
</ref>

=== Knowledge representation ===
<!-- This is linked to in the introduction -->
[[File:GFO taxonomy tree.png|thumb|An ontology represents knowledge as a set of concepts within a domain and the relationships between those concepts.]]

Knowledge representation and [[Knowledge engineering|knowledge engineering]]<ref>
Knowledge representation and [[Knowledge engineering|knowledge engineering]]:
* {{Harvtxt|Russell|Norvig|2003|pp=260–266, 320–363}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=23–46, 69–81, 169–233, 235–277, 281–298, 319–345}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=227–243}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 17.1–17.4, 18}}
</ref> allow AI programs to answer questions intelligently and make deductions about real-world facts.

A representation of "what exists" is an [[Ontology (information science)|ontology]]: the set of objects, relations, concepts, and properties formally described so that software agents can interpret them.{{sfnp|Russell|Norvig|2003|pp=320–328}} The most general ontologies are called [[Upper ontology|upper ontologies]], which attempt to provide a foundation for all other knowledge and act as mediators between domain ontologies that cover specific knowledge about a particular knowledge domain (field of interest or area of concern). A truly intelligent program would also need access to commonsense knowledge, the set of facts that an average person knows. The [[Philosophy:Semantics|semantics]] of an ontology is typically represented in description logic, such as the [[Web Ontology Language]].<ref name="Representing categories and relations"/>

AI research has developed tools to represent specific domains, such as objects, properties, categories and relations between objects;<ref name="Representing categories and relations"/> situations, events, states and time;<ref name="Representing time"/> causes and effects;<ref name="Representing causation"/> knowledge about knowledge (what we know about what other people know); and <ref name="Representing knowledge about knowledge"/> default reasoning (things that humans assume are true until they are told differently and will remain true even when other facts are changing);<ref name="Default reasoning and non-monotonic logic"/>. Among the most difficult problems in AI are: the breadth of commonsense knowledge (the number of atomic facts that the average person knows is enormous);<ref name="Breadth of commonsense knowledge">
Breadth of commonsense knowledge:
* {{Harvtxt|Russell|Norvig|2003|p=21}},
* {{Harvtxt|Crevier|1993|pp=113–114}},
* {{Harvtxt|Moravec|1988|p=13}},
* {{Harvtxt|Lenat|Guha|1989|loc=Introduction}}
</ref> and the sub-symbolic form of most commonsense knowledge (much of what people know is not represented as "facts" or "statements" that they could express verbally).<ref name="Psychological evidence of sub-symbolic reasoning"/>

Formal knowledge representations are used in content-based indexing and retrieval,{{sfnp|Smoliar|Zhang|1994}} scene interpretation,{{sfnp|Neumann|Möller|2008}} clinical decision support,{{sfnp|Kuperman|Reichley|Bailey|2006}} knowledge discovery (mining "interesting" and actionable inferences from large databases),{{sfnp|McGarry|2005}} and other areas.{{sfnp|Bertini|Del Bimbo|Torniai|2006}}

=== Learning ===
<!-- This is linked to in the introduction -->
{{Main|Machine learning}}

Machine learning (ML), a fundamental concept of AI research since the field's inception,{{efn|[[Biography:Alan Turing|Alan Turing]] discussed the centrality of learning as early as 1950, in his classic paper "[[Computing Machinery and Intelligence]]".{{sfnp|Turing|1950}} In 1956, at the original Dartmouth AI summer conference, [[Biography:Ray Solomonoff|Ray Solomonoff]] wrote a report on unsupervised probabilistic machine learning: "An Inductive Inference Machine".{{sfnp|Solomonoff|1956}}
}} is the study of computer algorithms that improve automatically through experience.{{efn|This is a form of [[Biography:Tom M. Mitchell|Tom Mitchell]]'s widely quoted definition of machine learning: "A computer program is set to learn from an experience ''E'' with respect to some task ''T'' and some performance measure ''P'' if its performance on ''T'' as measured by ''P'' improves with experience ''E''."{{sfnp|Russell|Norvig|2003|pp=649–788}}}}

[[Unsupervised learning]] finds patterns in a stream of input.

[[Supervised learning]] requires a human to label the input data first, and comes in two main varieties: [[Statistical classification|classification]] and numerical [[Regression analysis|regression]]. Classification is used to determine what category something belongs in – the program sees a number of examples of things from several categories and will learn to classify new inputs. Regression is the attempt to produce a function that describes the relationship between inputs and outputs and predicts how the outputs should change as the inputs change. Both classifiers and regression learners can be viewed as "function approximators" trying to learn an unknown (possibly implicit) function; for example, a spam classifier can be viewed as learning a function that maps from the text of an email to one of two categories, "spam" or "not spam".<ref>
[[Machine learning|Learning]]:
* {{Harvtxt|Russell|Norvig|2003|pp=649–788}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=397–438}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=385–542}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 3.3, 10.3, 17.5, 20}}
</ref>

In [[Reinforcement learning|reinforcement learning]] the agent is rewarded for good responses and punished for bad ones. The agent classifies its responses to form a strategy for operating in its problem space.<ref>
[[Reinforcement learning]]:
* {{Harvtxt|Russell|Norvig|2003|pp=763–788}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=442–449}}
</ref>

[[Transfer learning]] is when the knowledge gained from one problem is applied to a new problem.{{sfnp|The Economist|2016}}

[[Computational learning theory]] can assess learners by [[Computational complexity|computational complexity]], by [[Sample complexity|sample complexity]] (how much data is required), or by other notions of optimization.{{sfnp|Jordan|Mitchell|2015}}

{{clear}}

=== Natural language processing ===
<!-- This is linked to in the introduction -->
{{Main|Natural language processing}}
[[File:ParseTree.svg|thumb| A [[Parse tree|parse tree]] represents the syntactic structure of a sentence according to some [[Formal grammar|formal grammar]].|left]]

Natural language processing (NLP)<ref>
[[Natural language processing]] (NLP):
* {{Harvtxt|Russell|Norvig|2003|pp=790–831}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=91–104}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=591–632}}
</ref>
allows machines to read and [[Natural-language understanding|understand]] human language. A sufficiently powerful natural language processing system would enable [[Natural-language user interface|natural-language user interface]]s and the acquisition of knowledge directly from human-written sources, such as newswire texts. Some straightforward applications of NLP include [[Information retrieval|information retrieval]], [[Question answering|question answering]] and [[Machine translation|machine translation]].<ref>
Applications of [[Natural language processing|NLP]]:
* {{Harvtxt|Russell|Norvig|2003|pp=840–857}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=623–630}}
</ref>

Symbolic AI used formal [[Philosophy:Syntax|syntax]] to translate the deep structure of sentences into [[Logic|logic]]. This failed to produce useful applications, due to the intractability of logic<ref name="Intractability" /> and the breadth of commonsense knowledge.<ref name="Breadth of commonsense knowledge"/> Modern statistical techniques include co-occurrence frequencies (how often one word appears near another), "Keyword spotting" (searching for a particular word to retrieve information), [[Transformer (machine learning model)|transformer]]-based [[Deep learning|deep learning]] (which finds patterns in text), and others.<ref>Modern statistical approaches to [[Natural language processing|NLP]]:
* {{Harvtxt|Cambria|White|2014}}
</ref> They have achieved acceptable accuracy at the page or paragraph level, and, by 2019, could generate coherent text.{{sfnp|Vincent|2019}}

=== Perception ===<!-- This is linked to in the introduction -->
{{Main|Machine perception|Computer vision|Speech recognition}}
[[File:Ääretuvastuse näide.png|thumb|upright=1.5|[[Feature detection (computer vision)|Feature detection]] (pictured: [[Edge detection|edge detection]]) helps AI compose informative abstract structures out of raw data.]]

Machine perception<ref>
[[Machine perception]]:
* {{Harvtxt|Russell|Norvig|2003|pp=537–581, 863–898}}
* {{Harvtxt|Nilsson|1998|loc=~chpt. 6}}
</ref>
is the ability to use input from sensors (such as cameras, microphones, wireless signals, and active [[Engineering:Lidar|lidar]], sonar, radar, and [[Engineering:Tactile sensor|tactile sensor]]s) to deduce aspects of the world. Applications include [[Speech recognition|speech recognition]],<ref>
[[Speech recognition]]:
* {{Harvtxt|Russell|Norvig|2003|pp=568–578}}
</ref>
[[Facial recognition system|facial recognition]], and object recognition.<ref>
Object recognition:
* {{Harvtxt|Russell|Norvig|2003|pp=885–892}}
</ref>
Computer vision is the ability to analyze visual input.<ref>
[[Computer vision]]:
* {{Harvtxt|Russell|Norvig|2003|pp=863–898}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 6}}
</ref>

=== Social intelligence ===
<!-- This is linked to in the introduction -->
[[File:Kismet robot at MIT Museum.jpg|thumb|[[Engineering:Kismet (robot)|Kismet]], a robot with rudimentary social skills{{sfnp|MIT AIL|2014}}]]

Affective computing is an interdisciplinary umbrella that comprises systems that recognize, interpret, process or simulate human [[Philosophy:Affect (psychology)|feeling, emotion and mood]].<ref>
Affective computing:
* {{Harvtxt|Thro|1993}}
* {{Harvtxt|Edelson|1991}}
* {{Harvtxt|Tao|Tan|2005}}
* {{Harvtxt|Scassellati|2002}}
</ref>
For example, some [[Engineering:Virtual assistant|virtual assistant]]s are programmed to speak conversationally or even to banter humorously; it makes them appear more sensitive to the emotional dynamics of human interaction, or to otherwise facilitate [[Human–computer interaction|human–computer interaction]].
However, this tends to give naïve users an unrealistic conception of how intelligent existing computer agents actually are.{{sfnp|Waddell|2018}} Moderate successes related to affective computing include textual [[Sentiment analysis|sentiment analysis]] and, more recently, [[Multimodal sentiment analysis|multimodal sentiment analysis]], wherein AI classifies the affects displayed by a videotaped subject.{{sfnp|Poria|Cambria|Bajpai |Hussain|2017}}

=== General intelligence ===
<!-- This is linked to in the introduction -->
{{Main|Artificial general intelligence}}

A machine with general intelligence can solve a wide variety of problems with breadth and versatility similar to human intelligence. There are several competing ideas about how to develop artificial general intelligence. [[Biography:Hans Moravec|Hans Moravec]] and [[Biography:Marvin Minsky|Marvin Minsky]] argue that work in different individual domains can be incorporated into an advanced [[Multi-agent system|multi-agent system]] or [[Cognitive architecture|cognitive architecture]] with general intelligence.<ref>
''The Society of Mind'':
* {{Harvtxt|Minsky|1986}}
Moravec's "golden spike":
* {{Harvtxt|Moravec|1988|p=20}}
[[Multi-agent system]]s, hybrid intelligent systems, [[Agent architecture|agent architecture]]s, [[Cognitive architecture|cognitive architecture]]:
* {{Harvtxt|Russell|Norvig|2003|pp=27, 932, 970–972}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 25}}
</ref>
[[Biography:Pedro Domingos|Pedro Domingos]] hopes that there is a conceptually straightforward, but mathematically difficult, "[[The Master Algorithm|master algorithm]]" that could lead to AGI.{{sfnp|Domingos|2015|loc=Chpt. 9}}
Others believe that anthropomorphic features like an [[Philosophy:Artificial brain|artificial brain]]<ref>
[[Philosophy:Artificial brain|Artificial brain]] as an approach to [[Artificial general intelligence|AGI]]:
* {{Harvtxt|Russell|Norvig|2003|p=957}}
* {{Harvtxt|Crevier|1993|pp=271 & 279}}
* {{Harvtxt|Goertzel ''et al.''|2010}}
A few of the people who make some form of the argument:
* {{Harvtxt|Moravec|1988|p=20}}
* {{Harvtxt|Kurzweil|2005|p=262}}
* {{Harvtxt|Hawkins|Blakeslee|2005}}
</ref>
or simulated [[Developmental robotics|child development]]{{efn|
[[Biography:Alan Turing|Alan Turing]] suggested in "[[Computing Machinery and Intelligence]]" that a "thinking machine" would need to be educated like a child.{{sfnp|Turing|1950}} [[Developmental robotics]] is a modern version of the idea.<ref name="Developmental robotics"/>
}}
will someday reach a critical point where general intelligence [[Philosophy:Emergence|emerges]].

== Tools ==
=== Search and optimization ===
{{Main|Search algorithm|Mathematical optimization|Evolutionary computation}}
AI can solve many problems by intelligently searching through many possible solutions.<ref>
[[Search algorithm]]s:
* {{Harvtxt|Russell|Norvig|2003|pp=59–189}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=113–163}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=79–164, 193–219}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 7–12}}
</ref> [[Applications of artificial intelligence#Deduction, reasoning, problem solving|Reasoning]] can be reduced to performing a search. For example, logical proof can be viewed as searching for a path that leads from [[Premise|premise]]s to [[Logical consequence|conclusions]], where each step is the application of an inference rule.<ref name="Logic as search">[[Forward chaining]], [[Backward chaining|backward chaining]], [[Horn clause]]s, and logical deduction as search:
* {{Harvtxt|Russell|Norvig|2003|pp=217–225, 280–294}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=~46–52}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=62–73}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 4.2, 7.2}}
</ref> [[Automated planning and scheduling|Planning]] algorithms search through trees of goals and subgoals, attempting to find a path to a target goal, a process called [[Means-ends analysis|means-ends analysis]].<ref name="Planning as search">
[[State space search]] and [[Automated planning and scheduling|planning]]:
* {{Harvtxt|Russell|Norvig|2003|pp=382–387}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=298–305}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 10.1–2}}
</ref> [[Engineering:Robotics|Robotics]] algorithms for moving limbs and grasping objects use [[Local search (optimization)|local searches]] in [[Physics:Configuration space|configuration space]].<ref name="Configuration space">Moving and [[Physics:Configuration space|configuration space]]:
* {{Harvtxt|Russell|Norvig|2003|pp=916–932}}
</ref>

Simple exhaustive searches<ref name="Uninformed search">Uninformed searches (breadth first search, [[Depth-first search|depth-first search]] and general [[State space search|state space search]]):
* {{Harvtxt|Russell|Norvig|2003|pp=59–93}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=113–132}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=79–121}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 8}}
</ref>
are rarely sufficient for most real-world problems: the [[Search algorithm|search space]] (the number of places to search) quickly grows to astronomical numbers. The result is a search that is too slow or never completes. The solution, for many problems, is to use "heuristics" or "rules of thumb" that prioritize choices in favor of those more likely to reach a goal and to do so in a shorter number of steps. In some search methodologies, heuristics can also serve to eliminate some choices unlikely to lead to a goal (called "pruning the [[Search tree|search tree]]"). Heuristics supply the program with a "best guess" for the path on which the solution lies.<ref name="Informed search">
[[Heuristic]] or informed searches (e.g., greedy [[Best-first search|best first]] and [[A* search algorithm|A*]]):
* {{Harvtxt|Russell|Norvig|2003|pp=94–109}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=pp. 132–147}}
* {{Harvtxt|Poole|Mackworth|2017|loc=Section 3.6}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=133–150}}
</ref>
Heuristics limit the search for solutions into a smaller sample size.{{sfnp|Tecuci|2012}}

[[File:ParticleSwarmArrowsAnimation.gif|thumb|A [[Particle swarm optimization|particle swarm]] seeking the global minimum]]
A very different kind of search came to prominence in the 1990s, based on the mathematical theory of optimization. For many problems, it is possible to begin the search with some form of a guess and then refine the guess incrementally until no more refinements can be made. These algorithms can be visualized as blind [[Hill climbing|hill climbing]]: we begin the search at a random point on the landscape, and then, by jumps or steps, we keep moving our guess uphill, until we reach the top. Other related optimization algorithms include [[Random optimization|random optimization]], [[Beam search|beam search]] and [[Metaheuristic#Local search vs. global search|metaheuristics]] like [[Simulated annealing|simulated annealing]].<ref name="Optimization search">Optimization searches:
* {{Harvtxt|Russell|Norvig|2003|pp=110–116,120–129}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=56–163}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=127–133}}
</ref> [[Evolutionary computation]] uses a form of optimization search. For example, they may begin with a population of organisms (the guesses) and then allow them to mutate and recombine, selecting only the fittest to survive each generation (refining the guesses). Classic evolutionary algorithms include genetic algorithms, [[Gene expression programming|gene expression programming]], and [[Genetic programming|genetic programming]].<ref name="Genetic programming">
[[Genetic programming]] and genetic algorithms:
* {{Harvtxt|Luger|Stubblefield|2004|pp=509–530}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 4.2}}
</ref> Alternatively, distributed search processes can coordinate via [[Swarm intelligence|swarm intelligence]] algorithms. Two popular swarm algorithms used in search are [[Particle swarm optimization|particle swarm optimization]] (inspired by bird [[Flocking (behavior)|flocking]]) and ant colony optimization (inspired by ant trails).<ref name="Society based learning">
[[Artificial life]] and society based learning:
* {{Harvtxt|Luger|Stubblefield|2004|pp=530–541}}
* {{Harvtxt|Merkle|Middendorf|2013}}
</ref>

=== Logic ===
{{Main|Logic programming|Automated reasoning}}
[[Logic]]<ref name="Logic">
[[Logic]]:
* {{Harvtxt|Russell|Norvig|2003|pp=194–310}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=35–77}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 13–16}}</ref>
is used for knowledge representation and problem-solving, but it can be applied to other problems as well. For example, the [[Satplan|satplan]] algorithm uses logic for [[Automated planning and scheduling|planning]]<ref name="Satplan">
[[Satplan]]:
* {{Harvtxt|Russell|Norvig|2003|pp=402–407}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=300–301}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 21}}
</ref>
and [[Inductive logic programming|inductive logic programming]] is a method for [[Machine learning|learning]].<ref name="Symbolic learning techniques">
Explanation based learning, relevance based learning, [[Inductive logic programming|inductive logic programming]], case based reasoning:
* {{Harvtxt|Russell|Norvig|2003|pp=678–710}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=414–416}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=~422–442}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 10.3, 17.5}}
</ref>

Several different forms of logic are used in AI research. Propositional logic<ref name="Propositional logic">
Propositional logic:
* {{Harvtxt|Russell|Norvig|2003|pp=204–233}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=45–50}}
* {{Harvtxt|Nilsson|1998|loc=chpt. 13}}</ref> involves [[Truth function|truth function]]s such as "or" and "not". [[First-order logic]]<ref name="First-order logic">[[First-order logic]] and features such as [[Equality (mathematics)|equality]]:
* {{Harvtxt|Russell|Norvig|2003|pp=240–310}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=268–275}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=50–62}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 15}}
</ref>
adds [[Philosophy:Quantifier (logic)|quantifiers]] and [[Predicate (mathematical logic)|predicates]] and can express facts about objects, their properties, and their relations with each other. [[Fuzzy logic]] assigns a "degree of truth" (between 0 and 1) to vague statements such as "Alice is old" (or rich, or tall, or hungry), that are too linguistically imprecise to be completely true or false.<ref name="Fuzzy logic">
[[Fuzzy logic]]:
* {{Harvtxt|Russell|Norvig|2003|pp=526–527}}
* {{Harvtxt|Scientific American|1999}}
</ref>
[[Default logic]]s, [[Philosophy:Non-monotonic logic|non-monotonic logic]]s and [[Philosophy:Circumscription (logic)|circumscription]] are forms of logic designed to help with default reasoning and the [[Qualification problem|qualification problem]].<ref name="Default reasoning and non-monotonic logic">
Default reasoning, [[Frame problem]], [[Default logic|default logic]], [[Philosophy:Non-monotonic logic|non-monotonic logic]]s, [[Philosophy:Circumscription (logic)|circumscription]], closed world assumption, [[Abductive reasoning|abduction]]:
* {{Harvtxt|Russell|Norvig|2003|pp=354–360}}
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=248–256, 323–335}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=335–363}}
* {{Harvtxt|Nilsson|1998|loc=~18.3.3}}
(Poole ''et al.'' places abduction under "default reasoning". Luger ''et al.'' places this under "uncertain reasoning").
</ref>
Several extensions of logic have been designed to handle specific domains of knowledge, such as [[Description logic|description logic]]s;<ref name="Representing categories and relations">
Representing categories and relations: [[Semantic network]]s, [[Description logic|description logic]]s, inheritance (including [[Frame (artificial intelligence)|frames]] and scripts):
* {{Harvtxt|Russell|Norvig|2003|pp=349–354}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=174–177}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=248–258}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 18.3}}
</ref>
[[Situation calculus|situation calculus]], [[Event calculus|event calculus]] and fluent calculus (for representing events and time);<ref name="Representing time">Representing events and time:[[Situation calculus]], [[Event calculus|event calculus]], fluent calculus (including solving the [[Frame problem|frame problem]]):
* {{Harvtxt|Russell|Norvig|2003|pp=328–341}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=281–298}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 18.2}}
</ref>
[[Causality#Causal calculus|causal calculus]];<ref name="Representing causation">
[[Causality#Causal calculus|Causal calculus]]:
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=335–337}}
</ref>
[[Belief revision|belief calculus (belief revision)]]; and [[Modal logic|modal logic]]s.<ref name="Representing knowledge about knowledge">
Representing knowledge about knowledge: Belief calculus, [[Modal logic|modal logic]]s:
* {{Harvtxt|Russell|Norvig|2003|pp=341–344}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=275–277}}
</ref>
Logics to model contradictory or inconsistent statements arising in multi-agent systems have also been designed, such as [[Philosophy:Paraconsistent logic|paraconsistent logic]]s.<ref name="AbeNakamatsu">{{cite book | title = Knowledge Processing and Decision Making in Agent-Based Systems | last1 = Abe | first1 = Jair Minoro | last2 = Nakamatsu | first2 = Kazumi | chapter = Multi-agent Systems and Paraconsistent Knowledge | series = Studies in Computational Intelligence | year = 2009 | volume = 170 | pages = 101–121 | publisher = Springer Berlin Heidelberg | issn = 1860-949X | eissn = 1860-9503 | doi = 10.1007/978-3-540-88049-3_5 | isbn = 978-3-540-88048-6 | url = https://link.springer.com/chapter/10.1007/978-3-540-88049-3_5 | access-date = 2 August 2022 | archive-date = 9 February 2023 | archive-url = https://web.archive.org/web/20230209200254/https://link.springer.com/chapter/10.1007/978-3-540-88049-3_5 | url-status = live }}</ref>

=== Probabilistic methods for uncertain reasoning ===
[[File:EM_Clustering_of_Old_Faithful_data.gif|thumb|Expectation-maximization clustering of Old Faithful eruption data starts from a random guess but then successfully converges on an accurate clustering of the two physically distinct modes of eruption.]]

Many problems in AI (including in reasoning, planning, learning, perception, and robotics) require the agent to operate with incomplete or uncertain information. AI researchers have devised a number of tools to solve these problems using methods from [[Probability|probability]] theory and economics.<ref>
Stochastic methods for uncertain reasoning:
* {{Harvtxt|Russell|Norvig|2003|pp=462–644}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=345–395}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=165–191, 333–381}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 19}}
</ref>
[[Bayesian network]]s<ref name="Bayesian networks">
[[Bayesian network]]s:
* {{Harvtxt|Russell|Norvig|2003|pp=492–523}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=361–381}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=~182–190, ≈363–379}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 19.3–4}}
</ref>
are a very general tool that can be used for various problems, including reasoning (using the [[Bayesian inference]] algorithm),{{efn|
Compared with symbolic logic, formal Bayesian inference is computationally expensive. For inference to be tractable, most observations must be conditionally independent of one another. AdSense uses a Bayesian network with over 300&nbsp;million edges to learn which ads to serve.{{sfnp|Domingos|2015|loc=chapter 6}}
}}<ref name="Bayesian inference">
[[Bayesian inference]] algorithm:
* {{Harvtxt|Russell|Norvig|2003|pp=504–519}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=361–381}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=~363–379}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 19.4 & 7}}
</ref>
[[Machine learning|learning]] (using the expectation-maximization algorithm),{{efn|Expectation-maximization, one of the most popular algorithms in machine learning, allows clustering in the presence of unknown latent variables.{{sfnp|Domingos|2015|p=210}}}}<ref name="Bayesian learning">
Bayesian learning and the expectation-maximization algorithm:
* {{Harvtxt|Russell|Norvig|2003|pp=712–724}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=424–433}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 20}}
* {{Harvtxt|Domingos|2015|p=210}}
</ref>
[[Automated planning and scheduling|planning]] (using decision networks)<ref name="Bayesian decision networks">Bayesian decision theory and Bayesian decision networks:
* {{Harvtxt|Russell|Norvig|2003|pp=597–600}}</ref> and [[Machine perception|perception]] (using [[Dynamic Bayesian network|dynamic Bayesian network]]s).<ref name="Stochastic temporal models">Stochastic temporal models:
* {{Harvtxt|Russell|Norvig|2003|pp=537–581}}
[[Dynamic Bayesian network]]s:
* {{Harvtxt|Russell|Norvig|2003|pp=551–557}}
[[Hidden Markov model]]:
* {{Harv|Russell|Norvig|2003|pp=549–551}}
[[Kalman filter]]s:
* {{Harvtxt|Russell|Norvig|2003|pp=551–557}}
</ref>
Probabilistic algorithms can also be used for filtering, prediction, smoothing and finding explanations for streams of data, helping [[Machine perception|perception]] systems to analyze processes that occur over time (e.g., [[Hidden Markov model|hidden Markov model]]s or [[Kalman filter]]s).<ref name="Stochastic temporal models" />

A key concept from the science of economics is "[[Utility|utility]]", a measure of how valuable something is to an intelligent agent. Precise mathematical tools have been developed that analyze how an agent can make choices and plan, using [[Decision theory|decision theory]], [[Decision analysis|decision analysis]],<ref name="Decisions theory and analysis">
[[Decision theory|decision theory]] and [[Decision analysis|decision analysis]]:
* {{Harvtxt|Russell|Norvig|2003|pp=584–597}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=381–394}}
</ref>
and information value theory.<ref name="Information value theory">
Information value theory:
* {{Harvtxt|Russell|Norvig|2003|pp=600–604}}</ref> These tools include models such as [[Markov decision process]]es,<ref name="Markov decision process">[[Markov decision process]]es and dynamic decision networks:
* {{Harvtxt|Russell|Norvig|2003|pp=613–631}}</ref> dynamic decision networks,<ref name="Stochastic temporal models" /> [[Game theory|game theory]] and [[Mechanism design|mechanism design]].<ref name="Game theory and mechanism design">[[Game theory]] and [[Mechanism design|mechanism design]]:
* {{Harvtxt|Russell|Norvig|2003|pp=631–643}}
</ref>

=== Classifiers and statistical learning methods ===
{{Main|Statistical classification|Machine learning}}
The simplest AI applications can be divided into two types: classifiers ("if shiny then diamond") and controllers ("if diamond then pick up"). Controllers do, however, also classify conditions before inferring actions, and therefore classification forms a central part of many AI systems. Classifiers are functions that use [[Pattern matching|pattern matching]] to determine the closest match. They can be tuned according to examples, making them very attractive for use in AI. These examples are known as observations or patterns. In supervised learning, each pattern belongs to a certain predefined class. A class is a decision that has to be made. All the observations combined with their class labels are known as a data set. When a new observation is received, that observation is classified based on previous experience.<ref>
Statistical learning methods and classifiers:
* {{Harvtxt|Russell|Norvig|2003|pp=712–754}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=453–541}}
</ref>

A classifier can be trained in various ways; there are many statistical and [[Machine learning|machine learning]] approaches.
The [[Decision tree learning|decision tree]] is the simplest and most widely used symbolic machine learning algorithm.<ref>
[[Alternating decision tree|Decision tree]]:
* {{Harvtxt|Domingos|2015|p=88}}
* {{Harvtxt|Russell|Norvig|2003|pp=653–664}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=403–408}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=408–417}}
</ref>
K-nearest neighbor algorithm was the most widely used analogical AI until the mid-1990s.<ref>
K-nearest neighbor algorithm:
* {{Harvtxt|Domingos|2015|p=187}}
* {{Harvtxt|Russell|Norvig|2003|pp=733–736}}</ref>
Kernel methods such as the [[Support vector machine|support vector machine]] (SVM) displaced k-nearest neighbor in the 1990s.<ref>
kernel methods such as the [[Support vector machine|support vector machine]]:
* {{Harvtxt|Domingos|2015|p=88}}
* {{Harvtxt|Russell|Norvig|2003|pp=749–752}}
Gaussian mixture model:
* {{Harvtxt|Russell|Norvig|2003|pp=725–727}}
</ref>
The [[Naive Bayes classifier|naive Bayes classifier]] is reportedly the "most widely used learner"{{sfnp|Domingos|2015|p=152}} at Google, due in part to its scalability.<ref>
[[Naive Bayes classifier]]:
* {{Harvtxt|Domingos|2015|p=152}}
* {{Harvtxt|Russell|Norvig|2003|p=718}}
</ref>
[[Artificial neural network|Neural network]]s are also used for classification.<ref name="Neural networks"/>

Classifier performance depends greatly on the characteristics of the data to be classified, such as the dataset size, distribution of samples across classes, dimensionality, and the level of noise. Model-based classifiers perform well if the assumed model is an extremely good fit for the actual data. Otherwise, if no matching model is available, and if accuracy (rather than speed or scalability) is the sole concern, conventional wisdom is that discriminative classifiers (especially SVM) tend to be more accurate than model-based classifiers such as "naive Bayes" on most practical data sets.<ref>
Classifier performance:

* {{Harvtxt|van der Walt|Bernard|2006}}
* {{Harvtxt|Russell|Norvig|2009|loc=18.12: Learning from Examples: Summary}}
</ref>

=== Artificial neural networks ===
{{Main|Artificial neural network|Connectionism}}
[[File:Artificial_neural_network.svg|left|thumb|A neural network is an interconnected group of nodes, akin to the vast network of [[Biology:Neuron|neuron]]s in the human brain.]]
[[Artificial neural network|Neural network]]s<ref name="Neural networks">
Neural networks:
* {{Harvtxt|Russell|Norvig|2003|pp=736–748}},
* {{Harvtxt|Poole|Mackworth|Goebel|1998|pp=408–414}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=453–505}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 3}}
* {{Harvtxt|Domingos|2015|loc=Chapter 4}}
</ref>
were inspired by the architecture of neurons in the human brain. A simple "neuron" ''N'' accepts input from other neurons, each of which, when activated (or "fired"), casts a weighted "vote" for or against whether neuron ''N'' should itself activate. Learning requires an algorithm to adjust these weights based on the training data; one simple algorithm (dubbed "fire together, wire together") is to increase the weight between two connected neurons when the activation of one triggers the successful activation of another. Neurons have a continuous spectrum of activation; in addition, neurons can process inputs in a nonlinear way rather than weighing straightforward votes.

Modern neural networks model complex relationships between inputs and outputs and [[Pattern recognition|find patterns]] in data. They can learn continuous functions and even digital logical operations. Neural networks can be viewed as a type of [[Mathematical optimization|mathematical optimization]] – they perform [[Gradient descent|gradient descent]] on a multi-dimensional topology that was created by [[Machine learning|training]] the network. The most common training technique is the [[Backpropagation|backpropagation]] algorithm.<ref name="Backpropagation">
[[Backpropagation]]:
* {{Harvtxt|Russell|Norvig|2003|pp=744–748}},
* {{Harvtxt|Luger|Stubblefield|2004|pp=467–474}},
* {{Harvtxt|Nilsson|1998|loc=chpt. 3.3}}
[[Biography:Paul Werbos|Paul Werbos]]' introduction of backpropagation to AI:
* {{Harvtxt|Werbos|1974}}; {{Harvtxt|Werbos|1982}}
[[Automatic differentiation]], an essential precursor:
* {{Harvtxt|Linnainmaa|1970}}; {{Harvtxt|Griewank|2012}}
</ref>
Other [[Machine learning|learning]] techniques for neural networks are Hebbian learning ("fire together, wire together"), GMDH or [[Competitive learning|competitive learning]].<ref name="Learning in neural networks">
[[Competitive learning]], [[Hebbian theory|Hebbian]] coincidence learning, [[Hopfield network]]s and attractor networks:
* {{Harvtxt|Luger|Stubblefield|2004|pp=474–505}}
</ref>

The main categories of networks are acyclic or [[Feedforward neural network|feedforward neural network]]s (where the signal passes in only one direction) and [[Recurrent neural network|recurrent neural network]]s (which allow feedback and short-term memories of previous input events). Among the most popular feedforward networks are [[Perceptron|perceptron]]s, multi-layer perceptrons and radial basis networks.<ref name="Feedforward neural networks">
[[Feedforward neural network]]s, [[Perceptron|perceptron]]s and radial basis networks:
* {{Harvtxt|Russell|Norvig|2003|pp=739–748, 758}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=458–467}}
</ref>

{{Clear}}

==== Deep learning ====
[[File:Deep Learning.jpg|alt=Representing Images on Multiple Layers of Abstraction in Deep Learning|thumb|Representing images on multiple layers of abstraction in deep learning{{sfnp|Schulz| Behnke |2012}}]]
[[Deep learning]]<ref>
[[Deep learning]]:
* {{Harvtxt|Goodfellow|Bengio|Courville|2016}}
* {{Harvtxt|Hinton ''et al.''|2016}}
* {{Harvtxt|Schmidhuber|2015}}
</ref>
uses several layers of neurons between the network's inputs and outputs. The multiple layers can progressively extract higher-level features from the raw input. For example, in [[Image processing|image processing]], lower layers may identify edges, while higher layers may identify the concepts relevant to a human such as digits or letters or faces.{{sfnp|Deng|Yu|2014|pp=199–200}} Deep learning has drastically improved the performance of programs in many important subfields of artificial intelligence, including [[Computer vision|computer vision]], [[Speech recognition|speech recognition]], image classification{{sfnp|Ciresan|Meier|Schmidhuber|2012}} and others.

Deep learning often uses [[Convolutional neural network|convolutional neural network]]s for many or all of its layers. In a convolutional layer, each neuron receives input from only a restricted area of the previous layer called the neuron's [[Biology:Receptive field|receptive field]]. This can substantially reduce the number of weighted connections between neurons,{{sfnp|Habibi|2017}} and creates a hierarchy similar to the organization of the animal visual cortex.{{sfnp|Fukushima|2007}}

In a [[Recurrent neural network|recurrent neural network]] (RNN) the signal will propagate through a layer more than once;<ref>
Recurrent neural networks, Hopfield nets:
* {{Harvtxt|Russell|Norvig|2003|p=758}}
* {{Harvtxt|Luger|Stubblefield|2004|pp=474–505}}
* {{Harvtxt|Schmidhuber|2015}}
</ref>
thus, an RNN is an example of deep learning.{{sfnp|Schmidhuber|2015}}
RNNs can be trained by [[Gradient descent|gradient descent]],<ref><!-- should this be "backpropagation, which performs a [[Gradient descent|gradient descent]] on the network"? -->
{{Harvtxt|Werbos|1988}};
{{Harvtxt|Robinson|Fallside|1987}};
{{Harvtxt|Williams|Zipser|1994}}
</ref>
however long-term gradients which are back-propagated can "vanish" (that is, they can tend to zero) or "explode" (that is, they can tend to infinity), known as the [[Vanishing gradient problem|vanishing gradient problem]].<ref>
{{Harvtxt|Goodfellow|Bengio|Courville|2016}};
{{Harvtxt| Hochreiter|1991}}
</ref>
The long short term memory (LSTM) technique can prevent this in most cases.<ref>{{Harvtxt|Hochreiter| Schmidhuber|1997}}; {{Harvtxt| Gers | Schraudolph| Schraudolph |2002}}</ref>

=== Specialized languages and hardware ===
Specialized languages for artificial intelligence have been developed, such as [[Lisp (programming language)|Lisp]], [[Prolog]], [[Software:TensorFlow|TensorFlow]] and many others. Hardware developed for AI includes [[AI accelerator]]s and neuromorphic computing. By 2019, [[Graphics processing unit|graphics processing unit]]s (GPUs), often with AI-specific enhancements, had displaced [[Central processing unit|central processing unit]] (CPUs) as the dominant means to train large-scale commercial cloud AI.<ref>{{cite news |last=Kobielus |first=James |date=27 November 2019 |url=https://www.informationweek.com/ai-or-machine-learning/gpus-continue-to-dominate-the-ai-accelerator-market-for-now |title=GPUs Continue to Dominate the AI Accelerator Market for Now |work=InformationWeek |language=en |access-date=11 June 2020}}</ref>

== Applications ==
{{Main|Applications of artificial intelligence}}
[[File:Joseph Ayerle portrait of Ornella Muti (detail), calculated by Artificial Intelligence (AI) technology.jpg|thumb|For this 2018 project of the artist Joseph Ayerle the AI had to learn the typical patterns in the colors and brushstrokes of Renaissance painter Raphael. The portrait shows the face of the actress Ornella Muti, "painted" by AI in the style of Raphael.]]

AI is relevant to any intellectual task.{{sfnp|Russell|Norvig|2009|p=1}} Modern artificial intelligence techniques are pervasive and are too numerous to list here.{{sfnp|European Commission|2020|p=1}} Frequently, when a technique reaches mainstream use, it is no longer considered artificial intelligence; this phenomenon is described as the [[Philosophy:AI effect|AI effect]].{{sfnp|CNN|2006}}

<!-- High-profile applications  -->
In the 2010s, AI applications were at the heart of the most commercially successful areas of computing, and have become a ubiquitous feature of daily life. AI is used in search engines (such as [[Software:Google Search|Google Search]]),
[[Finance:Targeted advertising|targeting online advertisements]],<ref>
Targeted advertising:
* {{Harvtxt|Russell|Norvig|2009|p=1}}
* {{Harvtxt|Economist|2016}}
* {{Harvtxt|Lohr|2016}}</ref> [[Recommender system|recommendation systems]] (offered by [[Company:Netflix|Netflix]], [[Company:YouTube|YouTube]] or [[Company:Amazon|Amazon]]), driving [[Internet traffic|internet traffic]],{{sfnp|Lohr|2016}}{{sfnp|Smith|2016}} [[Software:Marketing and artificial intelligence|targeted advertising]] (AdSense, [[Company:Facebook|Facebook]]), [[Engineering:Virtual assistant|virtual assistant]]s (such as [[Software:Siri|Siri]] or [[Software:Amazon Alexa|Alexa]]),{{sfnp|Rowinski|2013}} autonomous vehicles (including [[Engineering:Unmanned aerial vehicle|drones]], [[Engineering:Advanced driver-assistance system|ADAS]] and self-driving cars), [[Machine translation|automatic language translation]] ([[Software:Microsoft Translator|Microsoft Translator]], [[Software:Google Translate|Google Translate]]), [[Facial recognition system|facial recognition]] (Apple's [[Face ID]] or [[Company:Microsoft|Microsoft]]'s [[Software:DeepFace|DeepFace]]), [[Automatic image annotation|image labeling]] (used by [[Company:Facebook|Facebook]], Apple's [[Software:IPhoto|iPhoto]] and [[Software:TikTok|TikTok]]), spam filtering and [[Chatbot|chatbot]]s (such as [[Software:ChatGPT|ChatGPT]]).

<!-- Uses throughout industry and academia -->
There are also thousands of successful AI applications used to solve problems for specific industries or institutions. A few examples are [[Engineering:Energy storage|energy storage]],{{sfnp|Frangoul|2019}} [[Deepfake|deepfake]]s,{{sfnp|Brown|2019}} medical diagnosis, military logistics, foreign policy,<ref>{{Cite web |date=2023-03-25 |title=Artificial intelligence, immune to fear or favour, is helping to make China's foreign policy {{!}} South China Morning Post |url=https://www.scmp.com/news/china/society/article/2157223/artificial-intelligence-immune-fear-or-favour-helping-make |access-date=2023-03-26 |archive-url=https://web.archive.org/web/20230325224424/https://www.scmp.com/news/china/society/article/2157223/artificial-intelligence-immune-fear-or-favour-helping-make |archive-date=25 March 2023 }}</ref> or supply chain management.

<!-- Notable experimental applications -->
Game playing has been a test of AI's strength since the 1950s. Deep Blue became the first computer chess-playing system to beat a reigning world chess champion, Garry Kasparov, on 11 May 1997.{{sfnp|McCorduck|2004|pp=480–483}} In 2011, in a ''[[Software:Jeopardy!|Jeopardy!]]'' quiz show exhibition match, [[Company:IBM|IBM]]'s question answering system, Watson, defeated the two greatest ''Jeopardy!'' champions, Brad Rutter and Ken Jennings, by a significant margin.{{sfnp|Markoff|2011}}
In March 2016, [[Software:AlphaGo|AlphaGo]] won 4 out of 5 games of [[Software:Go (game)|Go]] in a match with Go champion Lee Sedol, becoming the first [[Software:Computer Go|computer Go]]-playing system to beat a professional Go player without handicaps.<ref>
{{Harvtxt|Google|2016}}; {{Harvtxt|BBC|2016}}</ref> Other programs handle imperfect-information games; such as for [[Engineering:Poker|poker]] at a superhuman level, Pluribus{{efn|The Smithsonian reports: "Pluribus has bested poker pros in a series of six-player no-limit Texas Hold'em games, reaching a milestone in artificial intelligence research. It is the first bot to beat humans in a complex multiplayer competition."{{sfnp|Solly|2019}}}} and Cepheus.{{sfnp|Bowling|Burch|Johanson|Tammelin|2015}} [[Company:DeepMind|DeepMind]] in the 2010s developed a "generalized artificial intelligence" that could learn many diverse [[Engineering:Atari 2600|Atari]] games on its own.{{sfnp|Sample|2017}}

DeepMind's AlphaFold 2 (2020) demonstrated the ability to approximate, in hours rather than months, the 3D structure of a protein.{{sfnp|Heath|2020}} Other applications predict the result of judicial decisions,{{sfnp|Aletras ''et al.''|2016}} create art (such as poetry or painting) and prove mathematical theorems.

Generative AI gained widespread prominence in the 2020s. "[[Large language model]]" systems such as [[Software:GPT-3|GPT-3]] (2020), with 175 billion parameters, matched human performance on pre-existing benchmarks, albeit without the system attaining a commonsense understanding of the contents of the benchmarks.{{sfnp|Anadiotis|2020}} A Pew Research poll conducted in March 2023 found that 14% of Americans adults had tried [[Software:ChatGPT|ChatGPT]], a large language model fine-tuned using [[Reinforcement learning from human feedback|reinforcement learning from human feedback]].<ref>{{cite news |last1=Vogels |first1=Emily A. |title=A majority of Americans have heard of ChatGPT, but few have tried it themselves |url=https://www.pewresearch.org/short-reads/2023/05/24/a-majority-of-americans-have-heard-of-chatgpt-but-few-have-tried-it-themselves/ |access-date=15 June 2023 |work=Pew Research Center |date=24 May 2023 |archive-date=June 8, 2023 |archive-url=https://web.archive.org/web/20230608181200/https://www.pewresearch.org/short-reads/2023/05/24/a-majority-of-americans-have-heard-of-chatgpt-but-few-have-tried-it-themselves/ |url-status=live }}</ref>  While estimates varied wildly, [[Company:Goldman Sachs|Goldman Sachs]] suggested in 2023 that generative language AI could increase global GDP by 7% in the next ten years.<ref>{{cite news |title=Your job is (probably) safe from artificial intelligence |url=https://www.economist.com/finance-and-economics/2023/05/07/your-job-is-probably-safe-from-artificial-intelligence |access-date=18 June 2023 |work=The Economist |date=7 May 2023}}</ref><ref>{{cite web |title=Generative AI Could Raise Global GDP by 7% |url=https://www.goldmansachs.com/intelligence/pages/generative-ai-could-raise-global-gdp-by-7-percent.html |website=Goldman Sachs |access-date=18 June 2023}}</ref>

In 2023, the increasing realism and ease-of-use of AI-based text-to-image generators such as [[Software:Midjourney|Midjourney]], [[Software:DALL-E|DALL-E]], and [[Software:Stable Diffusion|Stable Diffusion]]<ref>{{Cite web |last=Verma |first=Pranshu |last2=Schaul |first2=Kevin |title=See why AI like ChatGPT has gotten so good, so fast |url=https://www.washingtonpost.com/business/interactive/2023/artificial-intelligence-tech-rapid-advances/ |access-date=2023-05-28 |website=Washington Post |language=en}}</ref><ref>{{Cite web |date=2023-04-11 |title=Will AI-generated images create a new crisis for fact-checkers? Experts are not so sure |url=https://reutersinstitute.politics.ox.ac.uk/news/will-ai-generated-images-create-new-crisis-fact-checkers-experts-are-not-so-sure |access-date=2023-05-28 |website=Reuters Institute for the Study of Journalism |language=en}}</ref> sparked a trend of [[Viral phenomenon|viral]] AI-generated photos. Widespread attention was gained by a fake photo of Pope Francis wearing a white puffer coat,<ref>{{Cite web |last=Novak |first=Matt |title=That Viral Image Of Pope Francis Wearing A White Puffer Coat Is Totally Fake |url=https://www.forbes.com/sites/mattnovak/2023/03/26/that-viral-image-of-pope-francis-wearing-a-white-puffer-coat-is-totally-fake/ |access-date=2023-05-28 |website=Forbes |language=en}}</ref> the fictional arrest of Donald Trump,<ref>{{Cite web |date=2023-03-24 |title=Trump shares deepfake photo of himself praying as AI images of arrest spread online |url=https://www.independent.co.uk/news/world/americas/us-politics/donald-trump-ai-praying-photo-b2307178.html |access-date=2023-05-28 |website=The Independent |language=en}}</ref> and a hoax of an attack on the Pentagon,<ref>{{Cite news |last=Oremus |first=Will |last2=Harwell |first2=Drew |last3=Armus |first3=Teo |date=2023-05-22 |title=A tweet about a Pentagon explosion was fake. It still went viral. |language=en-US |work=Washington Post |url=https://www.washingtonpost.com/technology/2023/05/22/pentagon-explosion-ai-image-hoax/ |access-date=2023-05-28 |issn=0190-8286}}</ref> as well as the usage in professional creative arts.<ref>{{Cite web |last=Kolirin |first=Lianne |date=2023-04-18 |title=Artist rejects photo prize after AI-generated image wins award |url=https://www.cnn.com/style/article/ai-photo-win-sony-scli-intl/index.html |access-date=2023-05-28 |website=CNN |language=en}}</ref><ref>{{Cite news|date=2023-06-16|title=How Generative AI Can Augment Human Creativity|work=Harvard Business Review|url=https://hbr.org/2023/07/how-generative-ai-can-augment-human-creativity|access-date=2023-06-20|issn=0017-8012}}</ref>

<!-- Patents on A.I. -->

== Intellectual property ==
[[File:AI Patent families for functional application categories and sub categories.png|thumb|upright=1.35|AI patent families for functional application categories and sub categories. [[Computer vision|Computer vision]] represents 49 percent of patent families related to a functional application in 2016.]]

In 2019, WIPO reported that AI was the most prolific [[Philosophy:Emerging technologies|emerging technology]] in terms of the number of [[Social:Patent|patent]] applications and granted patents, the [[Engineering:Internet of things|Internet of things]] was estimated to be the largest in terms of market size. It was followed, again in market size, by big data technologies, robotics, AI, 3D printing and the fifth generation of mobile services (5G).<ref>{{Cite web |title=Intellectual Property and Frontier Technologies |url=https://www.wipo.int/about-ip/en/frontier_technologies/ |website=WIPO |access-date=30 March 2022 |archive-date=2 April 2022 |archive-url=https://web.archive.org/web/20220402064804/https://www.wipo.int/about-ip/en/frontier_technologies/ |url-status=live }}</ref> Since AI emerged in the 1950s, 340,000 AI-related patent applications were filed by innovators and 1.6 million scientific papers have been published by researchers, with the majority of all AI-related patent filings published since 2013. Companies represent 26 out of the top 30 AI patent applicants, with universities or public research organizations accounting for the remaining four.<ref name="auto">{{Cite web |date=2019 |title=WIPO Technology Trends 2019 – Artificial Intelligence |url=https://www.wipo.int/edocs/pubdocs/en/wipo_pub_1055.pdf |archive-url=https://ghostarchive.org/archive/20221009/https://www.wipo.int/edocs/pubdocs/en/wipo_pub_1055.pdf |archive-date=2022-10-09 |url-status=live |website=WIPO}}</ref> The ratio of scientific papers to inventions has significantly decreased from 8:1 in 2010 to 3:1 in 2016, which is attributed to be indicative of a shift from theoretical research to the use of AI technologies in commercial products and services. [[Machine learning]] is the dominant AI technique disclosed in patents and is included in more than one-third of all identified inventions (134,777 machine learning patents filed for a total of 167,038 AI patents filed in 2016), with [[Computer vision|computer vision]] being the most popular functional application. AI-related patents not only disclose AI techniques and applications, they often also refer to an application field or industry. Twenty application fields were identified in 2016 and included, in order of magnitude: telecommunications (15 percent), transportation (15 percent), life and medical sciences (12 percent), and personal devices, computing and human–computer interaction (11 percent). Other sectors included banking, entertainment, security, industry and manufacturing, agriculture, and networks (including social networks, smart cities and the Internet of things). IBM has the largest portfolio of AI patents with 8,290 patent applications, followed by Microsoft with 5,930 patent applications.<ref name="auto"/>

== Philosophy ==
{{Main|Philosophy:Philosophy of artificial intelligence}}

=== Defining artificial intelligence ===
{{Main|Turing test|Intelligent agent|Dartmouth workshop|Synthetic intelligence}}
[[Biography:Alan Turing|Alan Turing]] wrote in 1950 "I propose to consider the question 'can machines think'?"{{sfnp|Turing|1950|p=1}}
He advised changing the question from whether a machine "thinks", to "whether or not it is possible for machinery to show intelligent behaviour".{{sfnp|Turing|1950|p=1}}
He devised the Turing test, which measures the ability of a machine to simulate human conversation.<ref name="Turing test">
Turing's original publication of the [[Turing test]] in "Computing machinery and intelligence":
* {{Harvtxt|Turing|1950}}
Historical influence and philosophical implications:
* {{Harvtxt|Haugeland|1985|pp=6–9}}
* {{Harvtxt|Crevier|1993|p=24}}
* {{Harvtxt|McCorduck|2004|pp=70–71}}
* {{Harvtxt|Russell|Norvig|2021|pp=2 and 984}}
</ref> Since we can only observe the behavior of the machine, it does not matter if it is "actually" thinking or literally has a "mind". Turing notes that we can not determine these things about other people{{efn|See [[Philosophy:Problem of other minds|Problem of other minds]]}} but "it is usual to have a polite convention that everyone thinks"{{sfnp|Turing|1950|loc=Under "The Argument from Consciousness"}}

[[Biography:Stuart J. Russell|Russell]] and [[Biography:Peter Norvig|Norvig]] agree with Turing that AI must be defined in terms of "acting" and not "thinking".{{sfnp|Russell|Norvig|2021|loc=chpt. 2}} However, they are critical that the test compares machines to ''people''. "[[Earth:Aeronautics|Aeronautical engineering]] texts," they wrote, "do not define the goal of their field as making 'machines that fly so exactly like pigeons that they can fool other pigeons.{{' "}}{{sfnp|Russell|Norvig|2021|p=3}} AI founder [[Biography:John McCarthy (computer scientist)|John McCarthy]] agreed, writing that "Artificial intelligence is not, by definition, simulation of human intelligence".{{sfnp|Maker|2006}}

McCarthy defines intelligence as "the computational part of the ability to achieve goals in the world."{{sfn|McCarthy|1999}} Another AI founder, [[Biography:Marvin Minsky|Marvin Minsky]] similarly defines it as "the ability to solve hard problems".{{sfnp|Minsky|1986}} These definitions view intelligence in terms of well-defined problems with well-defined solutions, where both the difficulty of the problem and the performance of the program are direct measures of the "intelligence" of the machine—and no other philosophical discussion is required, or may not even be possible.

A definition that has also been adopted by Google<ref>{{Cite web |title=Artificial intelligence - Google Search |url=https://www.google.com/search?q=Artificial+intelligence |access-date=2022-11-05 |website=www.google.com |archive-date=1 December 2022 |archive-url=https://web.archive.org/web/20221201011828/https://www.google.com/search?q=artificial+intelligence |url-status=live }}</ref>{{better source needed|date=December 2022}} – major practitionary in the field of AI.
This definition stipulated the ability of systems to synthesize information as the manifestation of intelligence, similar to the way it is defined in biological intelligence.

=== Evaluating approaches to AI ===

No established unifying theory or [[Philosophy:Paradigm|paradigm]] has guided AI research for most of its history.{{efn
|Nils Nilsson wrote in 1983: "Simply put, there is wide disagreement in the field about what AI is all about."{{sfnp|Nilsson|1983|p=10}}}} The unprecedented success of statistical machine learning in the 2010s eclipsed all other approaches (so much so that some sources, especially in the business world, use the term "artificial intelligence" to mean "machine learning with neural networks"). This approach is mostly sub-symbolic, [[Neats and scruffies|neat]], [[Soft computing|soft]] and [[Artificial general intelligence|narrow]] (see below). Critics argue that these questions may have to be revisited by future generations of AI researchers.

====Symbolic AI and its limits====

Symbolic AI (or "[[GOFAI]]"){{sfnp|Haugeland|1985|pp=112–117}} simulated the high-level conscious reasoning that people use when they solve puzzles, express legal reasoning and do mathematics. They were highly successful at "intelligent" tasks such as algebra or IQ tests. In the 1960s, Newell and Simon proposed the physical symbol systems hypothesis: "A physical symbol system has the necessary and sufficient means of general intelligent action."<ref name="Physical symbol system hypothesis">
Physical symbol system hypothesis:
* {{Harvtxt|Newell|Simon|1976|p=116}}
Historical significance:
* {{Harvtxt|McCorduck|2004|p=153}}
* {{Harvtxt|Russell|Norvig|2003|p=18}}
</ref>

However, the symbolic approach failed on many tasks that humans solve easily, such as learning, recognizing an object or commonsense reasoning. Moravec's paradox is the discovery that high-level "intelligent" tasks were easy for AI, but low level "instinctive" tasks were extremely difficult.<ref>
[[Moravec's paradox]]:
* {{Harvtxt|Moravec|1988|pp=15–16}}
* {{Harvtxt|Minsky|1986|p=29}}
* {{Harvtxt|Pinker|2007|pp=190–91}}
</ref>
Philosopher [[Biography:Hubert Dreyfus|Hubert Dreyfus]] had argued since the 1960s that human expertise depends on unconscious instinct rather than conscious symbol manipulation, and on having a "feel" for the situation, rather than explicit symbolic knowledge.<ref name="Dreyfus' critique">
Dreyfus' critique of AI:
* {{Harvtxt|Dreyfus|1972}}
* {{Harvtxt |Dreyfus|Dreyfus|1986}}
Historical significance and philosophical implications:
* {{Harvtxt |Crevier|1993|pp=120–132}}
* {{Harvtxt |McCorduck|2004|pp=211–239}}
* {{Harvtxt |Russell|Norvig|2003|pp=950–952}}
* {{Harvtxt |Fearn|2007|loc=Chpt. 3}}
</ref>
Although his arguments had been ridiculed and ignored when they were first presented, eventually, AI research came to agree.{{efn|
Daniel Crevier wrote that "time has proven the accuracy and perceptiveness of some of Dreyfus's comments. Had he formulated them less aggressively, constructive actions they suggested might have been taken much earlier."{{sfnp|Crevier|1993|p=125}}
}}<ref name="Psychological evidence of sub-symbolic reasoning"/>

The issue is not resolved: sub-symbolic reasoning can make many of the same inscrutable mistakes that human intuition does, such as [[Algorithmic bias|algorithmic bias]]. Critics such as [[Biography:Noam Chomsky|Noam Chomsky]] argue continuing research into symbolic AI will still be necessary to attain general intelligence,{{sfnp|Langley|2011}}{{sfnp|Katz|2012}} in part because sub-symbolic AI is a move away from explainable AI: it can be difficult or impossible to understand why a modern statistical AI program made a particular decision. The emerging field of [[Neuro-symbolic AI|neuro-symbolic artificial intelligence]] attempts to bridge the two approaches.

==== Neat vs. scruffy ====
{{Main|Neats and scruffies}}
"Neats" hope that intelligent behavior is described using simple, elegant principles (such as [[Logic|logic]], optimization, or neural networks). "Scruffies" expect that it necessarily requires solving a large number of unrelated problems (especially in areas like common sense reasoning). This issue was actively discussed in the 70s and 80s,<ref name="Neats vs. scruffies">
Neats vs. scruffies, the historic debate:
* {{Harvtxt|McCorduck|2004|pp=421–424, 486–489}}
* {{Harvtxt|Crevier|1993|p=168}}
* {{Harvtxt|Nilsson|1983|pp=10–11}}
A classic example of the "scruffy" approach to intelligence:
* {{Harvtxt|Minsky|1986}}
A modern example of neat AI and its aspirations:
* {{Harvtxt|Domingos|2015}}
</ref>
but in the 1990s mathematical methods and solid scientific standards became the norm, a transition that Russell and Norvig termed "the victory of the neats".{{sfnp|Russell|Norvig|2003|pp=25–26}}

==== Soft vs. hard computing ====
{{Main|Soft computing}}
Finding a provably correct or optimal solution is intractable for many important problems.<ref name="Intractability"/> Soft computing is a set of techniques, including genetic algorithms, [[Fuzzy logic|fuzzy logic]] and neural networks, that are tolerant of imprecision, uncertainty, partial truth and approximation. Soft computing was introduced in the late 80s and most successful AI programs in the 21st century are examples of soft computing with neural networks.

==== Narrow vs. general AI ====
{{Main|Artificial general intelligence}}
AI researchers are divided as to whether to pursue the goals of artificial general intelligence and [[Philosophy:Superintelligence|superintelligence]] (general AI) directly or to solve as many specific problems as possible ([[Engineering:Weak artificial intelligence|narrow AI]]) in hopes these solutions will lead indirectly to the field's long-term goals.{{sfnp|Pennachin|Goertzel|2007}}{{sfnp|Roberts|2016}}
General intelligence is difficult to define and difficult to measure, and modern AI has had more verifiable successes by focusing on specific problems with specific solutions. The experimental sub-field of artificial general intelligence studies this area exclusively.

=== Machine consciousness, sentience and mind ===
{{Main|Philosophy:Philosophy of artificial intelligence|Artificial consciousness}}
The [[Philosophy:Philosophy of mind|philosophy of mind]] does not know whether a machine can have a [[Philosophy:Mind|mind]], [[Philosophy:Consciousness|consciousness]] and [[Philosophy:Philosophy of mind|mental states]], in the same sense that human beings do. This issue considers the internal experiences of the machine, rather than its external behavior. Mainstream AI research considers this issue irrelevant because it does not affect the goals of the field. [[Biography:Stuart J. Russell|Stuart Russell]] and [[Biography:Peter Norvig|Peter Norvig]] observe that most AI researchers "don't care about the [philosophy of AI] – as long as the program works, they don't care whether you call it a simulation of intelligence or real intelligence."{{sfnp|Russell|Norvig|2003|p=947}} However, the question has become central to the philosophy of mind. It is also typically the central question at issue in artificial intelligence in fiction.

==== Consciousness ====
{{Main|Philosophy:Hard problem of consciousness|Philosophy:Theory of mind}}
[[Biography:David Chalmers|David Chalmers]] identified two problems in understanding the mind, which he named the "hard" and "easy" problems of consciousness.{{sfnp|Chalmers|1995}} The easy problem is understanding how the brain processes signals, makes plans and controls behavior. The hard problem is explaining how this ''feels'' or why it should feel like anything at all, assuming we are right in thinking that it truly does feel like something (Dennett's consciousness illusionism says this is an illusion). Human [[Information processing|information processing]] is easy to explain, however, human subjective experience is difficult to explain. For example, it is easy to imagine a color-blind person who has learned to identify which objects in their field of view are red, but it is not clear what would be required for the person to ''know what red looks like''.{{sfnp|Dennett|1991}}

==== Computationalism and functionalism ====
{{Main|Computational theory of mind|Philosophy:Functionalism (philosophy of mind)|Philosophy:Chinese room}}
Computationalism is the position in the [[Philosophy:Philosophy of mind|philosophy of mind]] that the human mind is an information processing system and that thinking is a form of computing. Computationalism argues that the relationship between mind and body is similar or identical to the relationship between software and hardware and thus may be a solution to the mind-body problem. This philosophical position was inspired by the work of AI researchers and cognitive scientists in the 1960s and was originally proposed by philosophers [[Biography:Jerry Fodor|Jerry Fodor]] and [[Biography:Hilary Putnam|Hilary Putnam]].{{sfnp|Horst|2005}}

Philosopher [[Biography:John Searle|John Searle]] characterized this position as "strong AI": "The appropriately programmed computer with the right inputs and outputs would thereby have a mind in exactly the same sense human beings have minds."{{efn|name="Searle's strong AI"|
Searle presented this definition of "Strong AI" in 1999.{{sfnp|Searle|1999}} Searle's original formulation was "The appropriately programmed computer really is a mind, in the sense that computers given the right programs can be literally said to understand and have other cognitive states."{{sfnp|Searle|1980|p=1}} Strong AI is defined similarly by [[Biography:Stuart J. Russell|Russell]] and [[Biography:Peter Norvig|Norvig]]: "The assertion that machines could possibly act intelligently (or, perhaps better, act ''as if'' they were intelligent) is called the 'weak AI' hypothesis by philosophers, and the assertion that machines that do so are actually thinking (as opposed to simulating thinking) is called the 'strong AI' hypothesis."{{sfnp|Russell|Norvig|2003|p=947}}
}}
Searle counters this assertion with his Chinese room argument, which attempts to show that, even if a machine perfectly simulates human behavior, there is still no reason to suppose it also has a mind.<ref name="Chinese room">
Searle's [[Philosophy:Chinese room|Chinese room]] argument:
* {{Harvtxt|Searle|1980}}. Searle's original presentation of the thought experiment.
* {{Harvtxt|Searle|1999}}.
Discussion:
* {{Harvtxt|Russell|Norvig|2003|pp=958–960}}
* {{Harvtxt|McCorduck|2004|pp=443–445}}
* {{Harvtxt|Crevier|1993|pp=269–271}}
</ref>

==== Robot rights ====
If a machine has a mind and subjective experience, then it may also have [[Philosophy:Sentience|sentience]] (the ability to feel), and if so it could also ''suffer''; it has been argued that this could entitle it to certain rights.<ref>
Robot rights:
* {{Harvtxt|Russell|Norvig|2003|p=964}}
* {{Harvtxt|BBC|2006}}
* {{Harvtxt|Maschafilm|2010}} (the film Plug & Pray)
</ref>
Any hypothetical robot rights would lie on a spectrum with [[Philosophy:Animal rights|animal rights]] and human rights.{{sfnp|Evans|2015}}
This issue has been considered in fiction for centuries,{{sfnp|McCorduck|2004|pp=19–25}}
and is now being considered by, for example, California's [[Organization:Institute for the Future|Institute for the Future]]; however, critics argue that the discussion is premature.{{sfnp|Henderson|2007}}

==Future==
=== Superintelligence ===
{{Main|Philosophy:Superintelligence|Philosophy:Technological singularity|Philosophy:Transhumanism}}
<!-- SUPERINTELLIGENCE -->
A superintelligence is a hypothetical agent that would possess intelligence far surpassing that of the brightest and most gifted human mind.{{sfnp|Roberts|2016}}

<!-- SINGULARITY -->
If research into [[Artificial general intelligence|artificial general intelligence]] produced sufficiently intelligent software, it might be able to reprogram and improve itself. The improved software would be even better at improving itself, leading to [[Philosophy:Intelligence explosion|recursive self-improvement]].{{sfnp|Omohundro|2008}}
Its intelligence would increase exponentially in an [[Philosophy:Intelligence explosion|intelligence explosion]] and could dramatically surpass humans. Science fiction writer [[Biography:Vernor Vinge|Vernor Vinge]] named this scenario the "singularity".{{sfnp|Vinge|1993}}
Because it is difficult or impossible to know the limits of intelligence or the capabilities of superintelligent machines, the technological singularity is an occurrence beyond which events are unpredictable or even unfathomable.{{sfnp|Russell|Norvig|2003|p=963}}

<!-- TRANSHUMANISM -->
Robot designer [[Biography:Hans Moravec|Hans Moravec]], cyberneticist [[Biography:Kevin Warwick|Kevin Warwick]], and inventor [[Biography:Ray Kurzweil|Ray Kurzweil]] have predicted that humans and machines will merge in the future into [[Cyborg|cyborg]]s that are more capable and powerful than either. This idea, called transhumanism, has roots in [[Biography:Aldous Huxley|Aldous Huxley]] and Robert Ettinger.<ref>
[[Philosophy:Transhumanism|Transhumanism]]:
* {{Harvtxt|Moravec|1988}}
* {{Harvtxt|Kurzweil|2005}}
* {{Harvtxt|Russell|Norvig|2003|p=963}}
</ref>

<!-- EVOLUTION / HUMAN REPLACEMENT -->
[[Biography:Edward Fredkin|Edward Fredkin]] argues that "artificial intelligence is the next stage in evolution", an idea first proposed by Samuel Butler's "[[Darwin among the Machines]]" as far back as 1863, and expanded upon by [[Biography:George Dyson (science historian)|George Dyson]] in his book of the same name in 1998.<ref>
AI as evolution:
* [[Biography:Edward Fredkin|Edward Fredkin]] is quoted in {{Harvtxt|McCorduck|2004|p=401}}
* {{Harvtxt|Butler|1863}}
* {{Harvtxt|Dyson|1998}}
</ref>

=== Risks ===

==== Technological unemployment ====
{{Main|Workplace impact of artificial intelligence|Engineering:Technological unemployment}}
<!-- TOPIC: AMOUNT OF UNEMPLOYMENT -->
In the past, technology has tended to increase rather than reduce total employment, but economists acknowledge that "we're in uncharted territory" with AI.<ref>{{Harvtxt|Ford|Colvin|2015}};{{Harvtxt|McGaughey|2022}}</ref> A survey of economists showed disagreement about whether the increasing use of robots and AI will cause a substantial increase in long-term unemployment, but they generally agree that it could be a net benefit if [[Finance:Productivity|productivity]] gains are [[Philosophy:Redistribution of income and wealth|redistributed]].{{sfnp|IGM Chicago|2017}} Risk estimates vary; for example, in the 2010s Michael Osborne and Carl Benedikt Frey estimated 47% of U.S. jobs are at "high risk" of potential automation, while an OECD report classified only 9% of U.S. jobs as "high risk".{{efn|See table 4; 9% is both the OECD average and the US average.{{sfnp|Arntz|Gregory|Zierahn|2016|p=33}}}}<ref>{{Harvtxt|Lohr|2017}}; {{Harvtxt|Frey|Osborne|2017}}; {{Harvtxt|Arntz|Gregory|Zierahn|2016|p=33}}</ref> The methodology of speculating about future employment levels has been criticised as lacking evidential foundation, and for implying that technology (rather than social policy) creates unemployment (as opposed to redundancies).<ref>E McGaughey, 'Will Robots Automate Your Job Away? Full Employment, Basic Income, and Economic Democracy' (2022) [https://academic.oup.com/ilj/article/51/3/511/6321008 51(3) Industrial Law Journal 511–559]</ref>

<!-- TOPIC: WHICH JOBS ARE AT RISK? -->
Unlike previous waves of automation, many middle-class jobs may be eliminated by artificial intelligence; ''[[Finance:The Economist|The Economist]]'' stated in 2015 that "the worry that AI could do to white-collar jobs what steam power did to blue-collar ones during the Industrial Revolution" is "worth taking seriously".{{sfnp|Morgenstern|2015}} Jobs at extreme risk range from paralegals to fast food cooks, while job demand is likely to increase for care-related professions ranging from personal healthcare to the clergy.<ref>{{Harvtxt|Mahdawi|2017}}; {{Harvtxt|Thompson|2014}}</ref>

==== Bad actors and weaponized AI ====
{{Main|Lethal autonomous weapon|Artificial intelligence arms race|AI safety}}

AI provides a number of tools that are particularly useful for authoritarian governments: smart [[Spyware|spyware]], [[Facial recognition system|face recognition]] and [[Speaker recognition|voice recognition]] allow widespread surveillance; such surveillance allows [[Machine learning|machine learning]] to classify potential enemies of the state and can prevent them from hiding; [[Recommender system|recommendation systems]] can precisely target [[Social:Propaganda|propaganda]] and misinformation for maximum effect; deepfakes aid in producing misinformation; advanced AI can make [[Social:Technocracy|centralized decision making]] more competitive with liberal and decentralized systems such as markets.{{sfnp|Harari|2018}}

Terrorists, criminals and rogue states may use other forms of weaponized AI such as advanced digital warfare and [[Lethal autonomous weapon|lethal autonomous weapon]]s. By 2015, over fifty countries were reported to be researching battlefield robots.<ref>
Weaponized AI:
* {{harvtxt|Robitzski|2018}}
* {{harvtxt|Sainato|2015}}
</ref>

Machine-learning AI is also able to design tens of thousands of toxic molecules in a matter of hours.<ref>{{cite journal |last1=Urbina |first1=Fabio |last2=Lentzos |first2=Filippa |last3=Invernizzi |first3=Cédric |last4=Ekins |first4=Sean |title=Dual use of artificial-intelligence-powered drug discovery |journal=Nature Machine Intelligence |date=7 March 2022 |volume=4 |issue=3 |pages=189–191 |doi=10.1038/s42256-022-00465-9 |pmid=36211133 |pmc=9544280 |s2cid=247302391 }}</ref>

==== Algorithmic bias ====
{{Main|Algorithmic bias}}

AI programs can become biased after learning from real-world data. It is not typically introduced by the system designers but is learned by the program, and thus the programmers are often unaware that the bias exists.{{sfnp|CNA|2019}}
Bias can be inadvertently introduced by the way training data is selected.{{sfnp|Goffrey|2008|p=17}}
It can also [[Algorithmic bias#Emergent|emerge]] from correlations: AI is used to classify individuals into groups and then make predictions assuming that the individual will resemble other members of the group. In some cases, this assumption may be unfair.<ref>{{Harvtxt|Lipartito|2011|p=36}}; {{Harvtxt|Goodman|Flaxman|2017|p=6}}</ref> An example of this is COMPAS, a commercial program widely used by U.S. courts to assess the likelihood of a defendant becoming a recidivist. ProPublica claims that the COMPAS-assigned recidivism risk level of black defendants is far more likely to be overestimated than that of white defendants, despite the fact that the program was not told the races of the defendants.{{sfnp|Larson|Angwin|2016}}

[[Medicine:Health equity|Health equity]] issues may also be exacerbated when [[Many-to-many (data model)|many-to-many]] mapping is done without taking steps to ensure equity for populations at risk for bias. At this time equity-focused tools and regulations are not in place to ensure equity application representation and usage.<ref>{{Cite journal |last1=Berdahl |first1=Carl Thomas |last2=Baker |first2=Lawrence |last3=Mann |first3=Sean |last4=Osoba |first4=Osonde |last5=Girosi |first5=Federico |date=2023-02-07 |title=Strategies to Improve the Impact of Artificial Intelligence on Health Equity: Scoping Review |url=https://ai.jmir.org/2023/1/e42936 |journal=JMIR AI |language=en |volume=2 |pages=e42936 |doi=10.2196/42936 |s2cid=256681439 |issn=2817-1705 |access-date=21 February 2023 |archive-date=21 February 2023 |archive-url=https://web.archive.org/web/20230221145255/https://ai.jmir.org/2023/1/e42936/ |url-status=live }}</ref> Other examples where algorithmic bias can lead to unfair outcomes are when AI is used for [[Finance:Credit rating|credit rating]] or hiring.

At its 2022 Conference on Fairness, Accountability, and Transparency (ACM FAccT 2022) the [[Organization:Association for Computing Machinery|Association for Computing Machinery]], in Seoul, South Korea, presented and published findings recommending that until AI and robotics systems are demonstrated to be free of bias mistakes, they are unsafe and the use of self-learning neural networks trained on vast, unregulated sources of flawed internet data should be curtailed.<ref>Dockrill, Peter, ''[https://www.sciencealert.com/robots-with-flawed-ai-make-sexist-racist-and-toxic-decisions-experiment-shows Robots With Flawed AI Make Sexist And Racist Decisions, Experiment Shows] {{Webarchive|url=https://web.archive.org/web/20220627225827/https://www.sciencealert.com/robots-with-flawed-ai-make-sexist-racist-and-toxic-decisions-experiment-shows |date=27 June 2022 }}'', Science Alert, 27 June 2022</ref>

==== Existential risk ====
{{Main|Unsolved:Existential risk from artificial general intelligence|AI alignment|AI safety}}

Superintelligent AI may be able to improve itself to the point that humans could not control it. This could, as physicist [[Biography:Stephen Hawking|Stephen Hawking]] puts it, "[[Unsolved:Global catastrophic risk|spell the end of the human race]]".{{sfnp|Cellan-Jones|2014}} Philosopher [[Biography:Nick Bostrom|Nick Bostrom]] argues that sufficiently intelligent AI, if it chooses actions based on achieving some goal, will exhibit [[Instrumental convergence|convergent]] behavior such as acquiring resources or protecting itself from being shut down. If this AI's goals do not fully reflect humanity's, it might need to harm humanity to acquire more resources or prevent itself from being shut down, ultimately to better achieve its goal. He concludes that AI poses a risk to mankind, however humble or "friendly" its stated goals might be.<ref>{{Harvtxt|Bostrom|2014}}; {{Harvtxt|Müller|Bostrom|2014}}; {{Harvtxt|Bostrom|2015}}</ref>
Political scientist Charles T. Rubin argues that "any sufficiently advanced benevolence may be indistinguishable from malevolence." Humans should not assume machines or robots would treat us favorably because there is no ''a priori'' reason to believe that they would share our system of morality.{{sfnp|Rubin|2003}}

The opinion of experts and industry insiders is mixed, with sizable fractions both concerned and unconcerned by risk from eventual superhumanly-capable AI.{{sfnp|Müller|Bostrom|2014}}
Physicist [[Biography:Stephen Hawking|Stephen Hawking]], [[Company:Microsoft|Microsoft]] founder [[Biography:Bill Gates|Bill Gates]], and [[Company:SpaceX|SpaceX]] founder [[Biography:Elon Musk|Elon Musk]] have expressed concern about existential risk from AI.<ref>
Leaders' concerns about the existential risks of AI:
* {{harvtxt|Rawlinson|2015}}
* {{Harvtxt|Holley|2015}}
* {{Harvtxt|Gibbs|2014}}
* {{harvtxt|Sainato|2015}}
</ref> In 2023 AI pioneers [[Biography:Geoffrey Hinton|Geoffrey Hinton]], [[Biography:Yoshua Bengio|Yoshua Bengio]], [[Biography:Demis Hassabis|Demis Hassabis]], and Sam Altman joined others to sign an open letter stating "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war"; others such as [[Biography:Yann LeCun|Yann LeCun]] dismissed such concerns as unfounded.<ref>{{cite news |title=Artificial intelligence could lead to extinction, experts warn |url=https://www.bbc.com/news/uk-65746524 |access-date=18 June 2023 |work=BBC News |date=30 May 2023}}</ref>
</ref>
Mark Zuckerberg (CEO, Facebook) has said that artificial intelligence is helpful in its current form and will continue to assist humans.<ref>
Leaders who argue the benefits of AI outweigh the risks:
* {{Harvtxt| Thibodeau |2019}}
* {{Harvtxt| Bhardwaj |2018}}
</ref>
Other experts argue is that the risks are far enough in the future to not be worth researching,
or that humans will be valuable from the perspective of a superintelligent machine.<ref>
Arguments that AI is not an imminent risk:
* {{Harvtxt|Brooks|2014}}
* {{Harvtxt|Geist |2015}}
* {{Harvtxt|Madrigal|2015}}
* {{Harvtxt|Lee |2014}}
</ref>
[[Biography:Rodney Brooks|Rodney Brooks]], in particular, has said that "malevolent" AI is still centuries away.{{efn|[[Biography:Rodney Brooks|Rodney Brooks]] writes, "I think it is a mistake to be worrying about us developing malevolent AI anytime in the next few hundred years. I think the worry stems from a fundamental error in not distinguishing the difference between the very real recent advances in a particular aspect of AI and the enormity and complexity of building sentient volitional intelligence."{{sfnp|Brooks|2014}}}}

==== Copyright ====
In order to leverage as large a dataset as is feasible, generative AI is often trained on unlicensed copyrighted works, including in domains such as images or computer code; the output is then used under a rationale of "[[Social:Fair use|fair use]]". Experts disagree about how well, and under what circumstances, this rationale will hold up in courts of law; relevant factors may include "the purpose and character of the use of the copyrighted work" and "the effect upon the potential market for the copyrighted work".<ref>{{cite news |last1=Vincent |first1=James |title=The scary truth about AI copyright is nobody knows what will happen next |url=https://www.theverge.com/23444685/generative-ai-copyright-infringement-legal-fair-use-training-data |access-date=19 June 2023 |work=The Verge |date=15 November 2022}}</ref>

=== Ethical machines ===

Friendly AI are machines that have been designed from the beginning to minimize risks and to make choices that benefit humans. [[Biography:Eliezer Yudkowsky|Eliezer Yudkowsky]], who coined the term, argues that developing friendly AI should be a higher research priority: it may require a large investment and it must be completed before AI becomes an existential risk.{{sfnp|Yudkowsky|2008}}

Machines with intelligence have the potential to use their intelligence to make ethical decisions. The field of machine ethics provides machines with ethical principles and procedures for resolving ethical dilemmas.{{sfnp|Anderson|Anderson|2011}}
The field of machine ethics is also called computational morality,{{sfnp|Anderson|Anderson|2011}}
and was founded at an AAAI symposium in 2005.{{sfnp|AAAI|2014}}

Other approaches include Wendell Wallach's "artificial moral agents"{{sfnp|Wallach|2010}}
and [[Biography:Stuart J. Russell|Stuart J. Russell]]'s three principles for developing provably beneficial machines.{{sfnp|Russell|2019|p=173}}

=== Regulation ===
{{Main|Regulation of artificial intelligence|Regulation of algorithms|AI safety}}
The regulation of artificial intelligence is the development of public sector policies and laws for promoting and regulating artificial intelligence (AI); it is therefore related to the broader regulation of algorithms.<ref>
Regulation of AI to mitigate risks:
* {{Harvtxt|Berryhill|Heang|Clogher|McBride|2019}}
* {{Harvtxt|Barfield|Pagallo|2018}}
* {{Harvtxt|Iphofen|Kritikos|2019}}
* {{Harvtxt|Wirtz|Weyerer|Geyer|2018}}
* {{Harvtxt|Buiten|2019}}
</ref>
The regulatory and policy landscape for AI is an emerging issue in jurisdictions globally.{{sfnp|Law Library of Congress (U.S.). Global Legal Research Directorate|2019}} According to AI Index at Stanford, the annual number of AI-related laws passed in the 127 survey countries jumped from one passed in 2016 to 37 passed in 2022 alone.<ref name="verge corporate"/><ref>{{cite web |title=Artificial Intelligence Index Report 2023/Chapter 6: Policy and Governance |url=https://aiindex.stanford.edu/wp-content/uploads/2023/04/HAI_AI-Index-Report-2023_CHAPTER_6-1.pdf |publisher=AI Index |access-date=19 June 2023 |date=2023}}</ref>
Between 2016 and 2020, more than 30 countries adopted dedicated strategies for AI.{{sfnp|UNESCO|2021}}
Most EU member states had released national AI strategies, as had Canada, China, India, Japan, Mauritius, the Russian Federation, Saudi Arabia, United Arab Emirates, US and Vietnam. Others were in the process of elaborating their own AI strategy, including Bangladesh, Malaysia and Tunisia.{{sfnp|UNESCO|2021}}
The Global Partnership on Artificial Intelligence was launched in June 2020, stating a need for AI to be developed in accordance with human rights and democratic values, to ensure public confidence and trust in the technology.{{sfnp|UNESCO|2021}} [[Biography:Henry Kissinger|Henry Kissinger]], Eric Schmidt, and [[Biography:Daniel P. Huttenlocher|Daniel Huttenlocher]] published a joint statement in November 2021 calling for a government commission to regulate AI.<ref>{{Cite news|last=Kissinger|first=Henry|date=1 November 2021|title=The Challenge of Being Human in the Age of AI|work={{wipe|The Wall Street Journal}}|url=https://www.wsj.com/articles/being-human-artifical-intelligence-ai-chess-antibiotic-philosophy-ethics-bill-of-rights-11635795271|access-date=4 November 2021|archive-date=4 November 2021|archive-url=https://web.archive.org/web/20211104012825/https://www.wsj.com/articles/being-human-artifical-intelligence-ai-chess-antibiotic-philosophy-ethics-bill-of-rights-11635795271|url-status=live}}</ref> In 2023, OpenAI leaders published recommendations for the governance of superintelligence, which they believe may happen in less than 10 years.<ref>{{Cite web |title=Governance of superintelligence |url=https://openai.com/blog/governance-of-superintelligence |access-date=2023-05-27 |website=openai.com |language=en-US}}</ref>

In a 2022 [[Company:Ipsos|Ipsos]] survey, attitudes towards AI varied greatly by country; 78% of Chinese citizens, but only 35% of Americans, agreed that "products and services using AI have more benefits than drawbacks".<ref name="verge corporate">{{cite news |last1=Vincent |first1=James |title=AI is entering an era of corporate control |url=https://www.theverge.com/23667752/ai-progress-2023-report-stanford-corporate-control |access-date=19 June 2023 |work=The Verge |date=3 April 2023}}</ref> A 2023 {{wipe|Reuters}}/Ipsos poll found that 61% of Americans agree, and 22% disagree, that AI poses risks to humanity.<ref>{{cite news |last1=Edwards |first1=Benj |title=Poll: AI poses risk to humanity, according to majority of Americans |url=https://arstechnica.com/information-technology/2023/05/poll-61-of-americans-say-ai-threatens-humanitys-future/ |access-date=19 June 2023 |work=Ars Technica |date=17 May 2023 |language=en-us}}</ref> In a 2023 Fox News poll, 35% of Americans thought it "very important", and an additional 41% thought it "somewhat important", for the federal government to regulate AI, versus 13% responding "not very important" and 8% responding "not at all important".<ref>{{cite news |last1=Kasperowicz |first1=Peter |title=Regulate AI? GOP much more skeptical than Dems that government can do it right: poll |url=https://www.foxnews.com/politics/regulate-ai-gop-much-more-skeptical-than-dems-that-the-government-can-do-it-right-poll |access-date=19 June 2023 |work=Fox News |date=1 May 2023}}</ref><ref>{{cite web |title=Fox News Poll |url=https://static.foxnews.com/foxnews.com/content/uploads/2023/05/Fox_April-21-24-2023_Complete_National_Topline_May-1-Release.pdf |publisher=Fox News |access-date=19 June 2023 |date=2023}}</ref>

== In fiction ==
[[File:Capek play.jpg|thumb|The word "robot" itself was coined by Karel Čapek in his 1921 play ''R.U.R.'', the title standing for "Rossum's Universal Robots".]]

Thought-capable artificial beings have appeared as storytelling devices since antiquity,<ref name="AI in myth" />
and have been a persistent theme in science fiction.{{sfnp|McCorduck|2004|pp=340–400}}

A common trope in these works began with Mary Shelley's ''Frankenstein'', where a human creation becomes a threat to its masters. This includes such works as Arthur C. Clarke's and Stanley Kubrick's ''2001: A Space Odyssey'' (both 1968), with HAL 9000, the murderous computer in charge of the ''Discovery One'' spaceship, as well as ''The Terminator'' (1984) and ''The Matrix'' (1999). In contrast, the rare loyal robots such as Gort from ''The Day the Earth Stood Still'' (1951) and Bishop from ''Aliens'' (1986) are less prominent in popular culture.{{sfnp|Buttazzo|2001}}

[[Biography:Isaac Asimov|Isaac Asimov]] introduced the Three Laws of Robotics in many books and stories, most notably the "Multivac" series about a super-intelligent computer of the same name. Asimov's laws are often brought up during lay discussions of machine ethics;{{sfnp|Anderson|2008}}
while almost all artificial intelligence researchers are familiar with Asimov's laws through popular culture, they generally consider the laws useless for many reasons, one of which is their ambiguity.{{sfnp|McCauley|2007}}

Several works use AI to force us to confront the fundamental question of what makes us human, showing us artificial beings that have [[Philosophy:Sentience|the ability to feel]], and thus to suffer. This appears in Karel Čapek's ''R.U.R.'', the films ''[[A.I. Artificial Intelligence]]'' and ''Ex Machina'', as well as the novel ''Do Androids Dream of Electric Sheep?'', by [[Biography:Philip K. Dick|Philip K. Dick]]. Dick considers the idea that our understanding of human subjectivity is altered by technology created with artificial intelligence.{{sfnp|Galvan|1997}}

== See also ==
* {{Annotated link |AI safety}}
* {{Annotated link |AI alignment}}
* [[Medicine:Artificial intelligence in healthcare|Artificial intelligence in healthcare]] – Machine-learning algorithms and software in the analysis, presentation, and comprehension of complex medical and health care data
* {{Annotated link |Artificial intelligence arms race}}
* {{Annotated link |Behavior selection algorithm}}
* {{Annotated link |Business process automation}}
* {{Annotated link |Case-based reasoning}}
* {{Annotated link |Emergent algorithm}}
* {{Annotated link |Female gendering of AI technologies}}
* {{Annotated link |Glossary of artificial intelligence}}
* {{Annotated link |Operations research}}
* {{Annotated link |Robotic process automation}}
* {{Annotated link |Synthetic intelligence}}
* {{Annotated link |Universal basic income}}
* {{Annotated link |Weak artificial intelligence}}
* [[List of datasets for machine-learning research|Data sources]] – The list of data sources for study and research

== Explanatory notes ==
{{Notelist}}

== References ==
{{reflist}}

=== AI textbooks ===
These were the four the most widely used AI textbooks in 2008:
{{refbegin}}
* {{cite book|last1=Luger|first1=George|last2=Stubblefield|first2=William|year=2004|title=Artificial Intelligence: Structures and Strategies for Complex Problem Solving|publisher=Benjamin/Cummings|edition=5th|isbn=978-0-8053-4780-7|url=https://archive.org/details/artificialintell0000luge|url-access=registration|access-date=17 December 2019|archive-date=26 July 2020|archive-url=https://web.archive.org/web/20200726220613/https://archive.org/details/artificialintell0000luge|url-status=live}}
* {{cite book|last=Nilsson|first=Nils|year=1998|title=Artificial Intelligence: A New Synthesis|url=https://archive.org/details/artificialintell0000nils|url-access=registration|publisher=Morgan Kaufmann|isbn=978-1-55860-467-4|access-date=18 November 2019|archive-date=26 July 2020|archive-url=https://web.archive.org/web/20200726131654/https://archive.org/details/artificialintell0000nils|url-status=live}}
* {{Russell Norvig 2003}}.
* {{cite book|first1=David|last1=Poole|first2=Alan|last2=Mackworth|first3=Randy|last3=Goebel|year=1998|title=Computational Intelligence: A Logical Approach|publisher=Oxford University Press|location=New York|isbn=978-0-19-510270-3|url=https://archive.org/details/computationalint00pool|access-date=22 August 2020|archive-date=26 July 2020|archive-url=https://web.archive.org/web/20200726131436/https://archive.org/details/computationalint00pool|url-status=live}}

Later editions.
* {{Cite book|first1=Stuart J.|last1=Russell|first2=Peter|last2=Norvig|title=[[Artificial Intelligence: A Modern Approach| A Modern Approach]] <!-- | url = http://aima.cs.berkeley.edu/ -->|year=2009|edition=3rd|publisher=Prentice Hall|location=Upper Saddle River, New Jersey|isbn=978-0-13-604259-4|}}.
* {{cite book|last1=Poole|first1=David|url=http://artint.info/index.html|title=Artificial Intelligence: Foundations of Computational Agents|last2=Mackworth|first2=Alan|publisher=Cambridge University Press|year=2017|isbn=978-1-107-19539-4|edition=2nd|access-date=6 December 2017|archive-date=7 December 2017|archive-url=https://web.archive.org/web/20171207013855/http://artint.info/index.html|url-status=live}}

The two most widely used textbooks in 2021.[https://opensyllabus.org/result/field?id=Computer+Science Open Syllabus: Explorer] {{Webarchive|url=https://web.archive.org/web/20211007144248/https://opensyllabus.org/result/field?id=Computer+Science |date=7 October 2021 }}
* {{Cite book
|title=[[Artificial Intelligence: A Modern Approach| A Modern Approach]]
|year=2021
|edition=4th
|isbn=9780134610993
|lccn=20190474
|publisher=Pearson|location=Hoboken
}}
* {{Cite book
| title = Artificial Intelligence
| date = 1 January 2010
| isbn = 9780070087705
| language = en
| publisher = Mc Graw Hill India
| edition = 3rd
}}
{{refend}}

=== History of AI ===
{{refbegin}}
* {{Crevier 1993}}.
* {{McCorduck 2004}}.
* {{cite book
| title=The Brain Makers: Genius, Ego, And Greed In The Quest For Machines That Think
| publisher=Macmillan/SAMS |location=New York
| isbn= 978-0-672-30412-5
}}
* {{cite book
| title=The Quest for Artificial Intelligence: A History of Ideas and Achievements
| publisher=Cambridge University Press |location=New York
| isbn=978-0-521-12293-1
}}
* {{citation
| last = McGaughey
| first = E
| title = Will Robots Automate Your Job Away? Full Employment, Basic Income, and Economic Democracy
| year = 2022
| page = 51(3) Industrial Law Journal 511–559
| ssrn = 3044448
| url = https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3044448
| access-date = 27 May 2023
| archive-date = 
}}
{{refend}}

=== Other sources ===
<!-- TO BE ALPHABETIZED -->

{{Refbegin}}
* {{citation
| last = McCarthy
| first = John
| year = 1999
| title = What is AI?
| url = http://jmc.stanford.edu/artificial-intelligence/what-is-ai/index.html
| access-date = 4 December 2022
| archive-date = 4 December 2022
| archive-url = https://web.archive.org/web/20221204051737/http://jmc.stanford.edu/artificial-intelligence/what-is-ai/index.html
| url-status = live
}}
* {{citation
| first = P. J.
| last = Werbos
| title = Generalization of backpropagation with application to a recurrent gas market model
| journal = Neural Networks
| year = 1988
| volume = 1
| issue = 4
| pages = 339–356
| doi = 10.1016/0893-6080(88)90007-X
| url = https://zenodo.org/record/1258627
| access-date = 29 September 2021
| archive-date = 29 October 2021
| archive-url = https://web.archive.org/web/20211029180324/https://zenodo.org/record/1258627
| url-status = live
}}
* {{Cite journal |url=http://www.jmlr.org/papers/volume3/gers02a/gers02a.pdf |archive-url=https://ghostarchive.org/archive/20221009/http://www.jmlr.org/papers/volume3/gers02a/gers02a.pdf |archive-date=2022-10-09 |url-status=live |title=Learning Precise Timing with LSTM Recurrent Networks |last1=Gers |first1=Felix A. |last2=Schraudolph |first2=Nicol N. |journal=Journal of Machine Learning Research |volume=3 |access-date=13 June 2017 |last3=Schraudolph |first3=Jürgen |pages=115–143 |year=2002 }}
* {{cite journal
 |last1        = Deng
 |first1       = L.
 |last2        = Yu
 |first2       = D.
 |year         = 2014
 |title        = Deep Learning: Methods and Applications
 |journal      = Foundations and Trends in Signal Processing
 |volume       = 7
 |issue        = 3–4
 |pages        = 1–199
 |doi          = 10.1561/2000000039
 |url          = http://research.microsoft.com/pubs/209355/DeepLearning-NowPublishing-Vol7-SIG-039.pdf
 |access-date  = 18 October 2014
 |archive-date = 14 March 2016
 |archive-url  = https://web.archive.org/web/20160314152112/http://research.microsoft.com/pubs/209355/DeepLearning-NowPublishing-Vol7-SIG-039.pdf
 |url-status   = live
}}
* {{Cite journal|last1=Schulz|first1=Hannes|last2=Behnke|first2=Sven|date=1 November 2012|title=Deep Learning|journal=KI – Künstliche Intelligenz|language=en|volume=26|issue=4|pages=357–363|doi=10.1007/s13218-012-0198-z|s2cid=220523562|issn=1610-1987|url=https://www.researchgate.net/publication/230690795}}
* {{cite journal | last1 = Fukushima | first1 = K. | year = 2007 | title = Neocognitron | journal = Scholarpedia | volume = 2 | issue = 1| page = 1717 | doi=10.4249/scholarpedia.1717| bibcode = 2007SchpJ...2.1717F | doi-access = free }} was introduced by [[Biography:Kunihiko Fukushima|Kunihiko Fukushima]] in 1980.
* {{Cite book|title=Guide to convolutional neural networks : a practical application to traffic-sign detection and classification|last=Habibi|first=Aghdam, Hamed|others=Heravi, Elnaz Jahani|isbn=9783319575490|location=Cham, Switzerland|oclc=987790957|date = 30 May 2017}}
* {{Cite book |doi=10.1109/cvpr.2012.6248110 |isbn=978-1-4673-1228-8|arxiv=1202.2745|chapter=Multi-column deep neural networks for image classification|title=2012 IEEE Conference on Computer Vision and Pattern Recognition|pages=3642–3649|year=2012|last1=Ciresan|first1=D.|last2=Meier|first2=U.|last3=Schmidhuber|first3=J.|s2cid=2161592}}
* {{cite news|ref={{harvid|The Economist|2016}}|title=From not working to neural networking|newspaper=The Economist|date=2016|url=https://www.economist.com/news/special-report/21700756-artificial-intelligence-boom-based-old-idea-modern-twist-not|access-date=26 April 2018|url-status=live|archive-date=31 December 2016|archive-url=https://web.archive.org/web/20161231203934/https://www.economist.com/news/special-report/21700756-artificial-intelligence-boom-based-old-idea-modern-twist-not}}
* {{cite magazine
 |last1        = Thompson
 |first1       = Derek
 |title        = What Jobs Will the Robots Take?
 |date         = 23 January 2014
 |magazine     = The Atlantic
 |url          = https://www.theatlantic.com/business/archive/2014/01/what-jobs-will-the-robots-take/283239/
 |access-date  = 24 April 2018
 |archive-date = 24 April 2018
 |archive-url  = https://web.archive.org/web/20180424202435/https://www.theatlantic.com/business/archive/2014/01/what-jobs-will-the-robots-take/283239/
 |url-status   = live
}}
* {{cite journal|last1=Scassellati|first1=Brian|title=Theory of mind for a humanoid robot|journal=Autonomous Robots|volume=12|issue=1|year=2002|pages=13–24 |doi=10.1023/A:1013298507114|s2cid=1979315}}
* {{cite news|last1=Sample|first1=Ian|title=Google's DeepMind makes AI program that can learn like a human|url=https://www.theguardian.com/global/2017/mar/14/googles-deepmind-makes-ai-program-that-can-learn-like-a-human|access-date=26 April 2018|work=The Guardian|date=14 March 2017|archive-date=26 April 2018|archive-url=https://web.archive.org/web/20180426212908/https://www.theguardian.com/global/2017/mar/14/googles-deepmind-makes-ai-program-that-can-learn-like-a-human|url-status=live}}
* {{cite news |last1=Heath |first1=Nick |title=What is AI? Everything you need to know about Artificial Intelligence |url=https://www.zdnet.com/article/what-is-ai-everything-you-need-to-know-about-artificial-intelligence/ |access-date=1 March 2021 |work=ZDNet |date=11 December 2020 |language=en |archive-date=2 March 2021 |archive-url=https://web.archive.org/web/20210302205428/https://www.zdnet.com/article/what-is-ai-everything-you-need-to-know-about-artificial-intelligence/ |url-status=live }}
* {{Cite journal|last1=Bowling|first1=Michael|last2=Burch|first2=Neil|last3=Johanson|first3=Michael|last4=Tammelin|first4=Oskari|date=9 January 2015|title=Heads-up limit hold'em poker is solved|url=https://www.science.org/doi/10.1126/science.1259433|journal=Science|language=en|volume=347|issue=6218|pages=145–149|doi=10.1126/science.1259433|issn=0036-8075|pmid=25574016|bibcode=2015Sci...347..145B|s2cid=3796371|access-date=30 June 2022|archive-date=1 August 2022|archive-url=https://web.archive.org/web/20220801134446/https://www.science.org/doi/10.1126/science.1259433|url-status=live}}
* {{Cite web
|url=https://www.smithsonianmag.com/smart-news/poker-playing-ai-knows-when-hold-em-when-fold-em-180972643/
|title=This Poker-Playing A.I. Knows When to Hold 'Em and When to Fold 'Em
|first=Meilan
|last=Solly
|date=15 July 2019
|website=Smithsonian
|access-date=1 October 2021
|archive-date=26 September 2021
|archive-url=https://web.archive.org/web/20210926070851/https://www.smithsonianmag.com/smart-news/poker-playing-ai-knows-when-hold-em-when-fold-em-180972643/
|url-status=live
}}
* {{cite news|ref={{harvid|BBC|2016}}|title=Artificial intelligence: Google's AlphaGo beats Go master Lee Se-dol|url=https://www.bbc.com/news/technology-35785875|access-date=1 October 2016|work=BBC News|date=12 March 2016|url-status=live|archive-url=https://web.archive.org/web/20160826103910/http://www.bbc.com/news/technology-35785875|archive-date=26 August 2016}}
* {{cite web|url=http://readwrite.com/2013/01/15/virtual-personal-assistants-the-future-of-your-smartphone-infographic|title=Virtual Personal Assistants & The Future Of Your Smartphone [Infographic]|date=15 January 2013|last=Rowinski|first=Dan|work=ReadWrite|url-status=live|archive-url=https://web.archive.org/web/20151222083034/http://readwrite.com/2013/01/15/virtual-personal-assistants-the-future-of-your-smartphone-infographic|archive-date=22 December 2015}}
* {{cite journal |last1=Manyika |first1=James |date=2022 |title=Getting AI Right: Introductory Notes on AI & Society |url=https://www.amacad.org/publication/getting-ai-right-introductory-notes-ai-society |journal=Daedalus |volume=151 |issue=2 |pages=5–27 |doi=10.1162/daed_e_01897 |s2cid=248377878 |access-date=5 May 2022 |doi-access=free |archive-date=5 May 2022 |archive-url=https://web.archive.org/web/20220505183207/https://www.amacad.org/publication/getting-ai-right-introductory-notes-ai-society |url-status=live }}
* {{cite news
|last=Markoff
|first=John
|date=16 February 2011<!-- corrected 24 February 2011-->
|title=Computer Wins on 'Jeopardy!': Trivial, It's Not
|work=The New York Times
|url=https://www.nytimes.com/2011/02/17/science/17jeopardy-watson.html
|access-date=25 October 2014
|archive-date=22 October 2014
|archive-url=https://web.archive.org/web/20141022023202/http://www.nytimes.com/2011/02/17/science/17jeopardy-watson.html
|url-status=live
}}
* {{cite news |last1=Anadiotis |first1=George |title=The state of AI in 2020: Democratization, industrialization, and the way to artificial general intelligence |url=https://www.zdnet.com/article/the-state-of-ai-in-2020-democratization-industrialization-and-the-way-to-artificial-general-intelligence/ |access-date=1 March 2021 |work=ZDNet |date=1 October 2020 |language=en |archive-date=15 March 2021 |archive-url=https://web.archive.org/web/20210315103618/https://www.zdnet.com/article/the-state-of-ai-in-2020-democratization-industrialization-and-the-way-to-artificial-general-intelligence/ |url-status=live }}
* {{cite journal | ref={{harvid|Goertzel ''et al.''|2010}}
|last1=Goertzel |first1=Ben |last2=Lian |first2=Ruiting |last3=Arel |first3=Itamar |last4=de Garis |first4=Hugo |last5=Chen |first5=Shuo |title=A world survey of artificial brain projects, Part II: Biologically inspired cognitive architectures |journal=Neurocomputing |date=December 2010 |volume=74 |issue=1–3 |pages=30–49 |doi=10.1016/j.neucom.2010.08.012}}
* {{citation
| first1 = A. J. |last1 = Robinson
| first2 = F. | last2 = Fallside
| title = The utility driven dynamic error propagation network.
| work = Technical Report CUED/F-INFENG/TR.1
| publisher = Cambridge University Engineering Department
| year = 1987}}
* {{cite thesis
| first = Sepp
| last = Hochreiter
| year = 1991
| url = http://people.idsia.ch/~juergen/SeppHochreiter1991ThesisAdvisorSchmidhuber.pdf
| title = Untersuchungen zu dynamischen neuronalen Netzen
| degree = diploma
| institution = Institut f. Informatik, Technische Univ
| location = Munich
| access-date = 16 April 2016
| archive-date = 6 March 2015
| archive-url = https://web.archive.org/web/20150306075401/http://people.idsia.ch/~juergen/SeppHochreiter1991ThesisAdvisorSchmidhuber.pdf
| url-status = dead
}}
* {{citation
| first1 = R. J. | last1 = Williams
| first2 = D. | last2 = Zipser
| title = Gradient-based learning algorithms for recurrent networks and their computational complexity
| journal = Back-propagation: Theory, Architectures and Applications
| location = Hillsdale, NJ: Erlbaum
| year = 1994
}}
* {{citation
| title = Long Short-Term Memory
| year = 1997
| journal = Neural Computation| volume = 9 | issue = 8 | pages = 1735–1780
| doi = 10.1162/neco.1997.9.8.1735 | pmid = 9377276 | s2cid = 1915014 }}
* {{citation
| first1 = Ian
| last1 = Goodfellow
| first2 = Yoshua
| last2 = Bengio
| first3 = Aaron
| last3 = Courville
| year = 2016
| title = Deep Learning
| publisher = MIT Press.
| url = http://www.deeplearningbook.org/
| access-date = 12 November 2017
| archive-date = 16 April 2016
| archive-url = https://web.archive.org/web/20160416111010/http://www.deeplearningbook.org/
| url-status = dead
}}
* {{cite journal | ref = {{Harvid|Hinton ''et al.''|2016}}
|last1=Hinton|first1=G.
|last2=Deng|first2=L.|last3=Yu|first3=D.
|last4=Dahl|first4=G.|last5=Mohamed|first5=A.
|last6=Jaitly|first6=N.|last7=Senior|first7=A.
|last8=Vanhoucke|first8=V.|last9=Nguyen|first9=P.
|year=2012
|title=Deep Neural Networks for Acoustic Modeling in Speech Recognition – The shared views of four research groups
|journal=IEEE Signal Processing Magazine|volume=29|issue=6|pages=82–97
|bibcode=2012ISPM...29...82H|doi=10.1109/msp.2012.2205597|s2cid=206485943}}
* {{cite journal|last=Schmidhuber|first=J.|author-link=Jürgen Schmidhuber
|year=2015|title=Deep Learning in Neural Networks: An Overview|journal=Neural Networks|volume=61|pages=85–117|arxiv=1404.7828|doi=10.1016/j.neunet.2014.09.003|pmid=25462637|s2cid=11715509}}
* {{cite thesis
| title = The representation of the cumulative rounding error of an algorithm as a Taylor expansion of the local rounding errors.
| year = 1970
| language = Finnish
| publisher = Univ. Helsinki, 6–7.
}}|
* {{cite journal
| last = Griewank | first = Andreas
| title = Who Invented the Reverse Mode of Differentiation? Optimization Stories
| year = 2012
| journal =Documenta Matematica, Extra Volume ISMP
| pages = 389–400
}}
* {{cite thesis
| title = Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences
| year = 1974
| degree = Ph.D.
| publisher = [[Organization:Harvard University|Harvard University]]
}}
* {{cite journal
| last = Werbos
| first = Paul
| title = Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences
| series = Applications of advances in nonlinear sensitivity analysis
| year = 1982
| journal = System Modeling and Optimization
| publisher = Springer
| location = Berlin, Heidelberg
| url = http://werbos.com/Neural/SensitivityIFIPSeptember1981.pdf
| access-date = 16 April 2016
| archive-date = 14 April 2016
| archive-url = https://web.archive.org/web/20160414055503/http://werbos.com/Neural/SensitivityIFIPSeptember1981.pdf
| url-status = dead
}}
* {{cite news
| ref = {{harvid|Scientific American|1999}}
| title = What is 'fuzzy logic'? Are there computers that are inherently fuzzy and do not apply the usual binary logic?
| date = 21 October 1999
| language = en
| work = Scientific American
| url = https://www.scientificamerican.com/article/what-is-fuzzy-logic-are-t/
| access-date = 5 May 2018
| archive-date = 6 May 2018
| archive-url = https://web.archive.org/web/20180506035133/https://www.scientificamerican.com/article/what-is-fuzzy-logic-are-t/
| url-status = live
}}
* {{cite book
| last1 = Merkle | first1 = Daniel
| last2 = Middendorf | first2 = Martin
| date = 2013
| title = Search Methodologies: Introductory Tutorials in Optimization and Decision Support Techniques
| publisher=Springer Science & Business Media
| isbn=978-1-4614-6940-7
| editor1-last=Burke |editor1-first=Edmund K.
| language=en
| chapter=Swarm Intelligence|editor2-last=Kendall|editor2-first=Graham}}
* {{cite web|last1=van der Walt|first1=Christiaan|last2=Bernard|first2=Etienne|year=2006|title=Data characteristics that determine classifier performance|url=http://www.patternrecognition.co.za/publications/cvdwalt_data_characteristics_classifiers.pdf|url-status=dead|archive-url=https://web.archive.org/web/20090325194051/http://www.patternrecognition.co.za/publications/cvdwalt_data_characteristics_classifiers.pdf|archive-date=25 March 2009|access-date=5 August 2009}}
* {{cite book|last=Hutter|first=Marcus|year=2005|title=Universal Artificial Intelligence|isbn=978-3-540-22139-5|publisher=Springer|location=Berlin|title-link=AIXI}}
* {{cite web
 |first = J.
 |last = Howe
 |date = November 1994
 |title = Artificial Intelligence at Edinburgh University: a Perspective
 |url = http://www.inf.ed.ac.uk/about/AIhistory.html
 |access-date = 30 August 2007
 |archive-date = 15 May 2007
 |archive-url = https://web.archive.org/web/20070515072641/http://www.inf.ed.ac.uk/about/AIhistory.html
 |url-status = live
 }}
* {{Cite journal|last=Galvan|first=Jill|date=1 January 1997|title=Entering the Posthuman Collective in Philip K. Dick's "Do Androids Dream of Electric Sheep?"|journal=Science Fiction Studies|volume=24|issue=3|pages=413–429|jstor=4240644}}
* {{cite journal | last1 = McCauley | first1 = Lee | year = 2007 | title = AI armageddon and the three laws of robotics | journal = Ethics and Information Technology | volume = 9 | issue = 2| pages = 153–164 | doi=10.1007/s10676-007-9138-2| citeseerx = 10.1.1.85.8904| s2cid = 37272949 }}
* {{cite journal|last1=Buttazzo|first1=G.|title=Artificial consciousness: Utopia or real possibility?|journal=Computer|date=July 2001|volume=34|issue=7|pages=24–30|doi=10.1109/2.933500}}
* {{cite journal
| last = Anderson | first = Susan Leigh
| title = Asimov's "three laws of robotics" and machine metaethics.
| year = 2008
| journal = AI & Society | volume = 22 | issue = 4 | pages = 477–493| doi = 10.1007/s00146-007-0094-5
| s2cid = 1809459
}}
* {{citation
| last = Yudkowsky
| first = E
| title = Artificial Intelligence as a Positive and Negative Factor in Global Risk
| year = 2008
| url = http://intelligence.org/files/AIPosNegFactor.pdf
| work = Global Catastrophic Risks
| publisher = Oxford University Press, 2008
| bibcode = 2008gcr..book..303Y
| access-date = 24 September 2021
| archive-date = 19 October 2013
| archive-url = https://web.archive.org/web/20131019182403/http://intelligence.org/files/AIPosNegFactor.pdf
| url-status = live
}}
* {{Cite web|author=IGM Chicago|title=Robots and Artificial Intelligence|date=30 June 2017|url=http://www.igmchicago.org/surveys/robots-and-artificial-intelligence|website=www.igmchicago.org|access-date=3 July 2019|archive-date=1 May 2019|archive-url=https://web.archive.org/web/20190501114826/http://www.igmchicago.org/surveys/robots-and-artificial-intelligence|url-status=live}}
* {{cite news|last1=Lohr|first1=Steve|date=2017|title=Robots Will Take Jobs, but Not as Fast as Some Fear, New Report Says|work=The New York Times|url=https://www.nytimes.com/2017/01/12/technology/robots-will-take-jobs-but-not-as-fast-as-some-fear-new-report-says.html|access-date=13 January 2018|archive-date=14 January 2018|archive-url=https://web.archive.org/web/20180114073704/https://www.nytimes.com/2017/01/12/technology/robots-will-take-jobs-but-not-as-fast-as-some-fear-new-report-says.html|url-status=live}}
* {{Cite journal|last1=Frey|first1=Carl Benedikt|last2=Osborne|first2=Michael A|date=1 January 2017|title=The future of employment: How susceptible are jobs to computerisation?|journal=Technological Forecasting and Social Change|volume=114|pages=254–280|citeseerx=10.1.1.395.416|doi=10.1016/j.techfore.2016.08.019|issn=0040-1625}}
* {{citation
| last1 = Arntz | first1 = Melanie
| last2 = Gregory | first2 = Terry
| last3 = Zierahn | first3 = Ulrich
| title = The risk of automation for jobs in OECD countries: A comparative analysis
| year = 2016
| work = OECD Social, Employment, and Migration Working Papers 189
}}
* {{cite news
| last = Morgenstern
| first = Michael
| date = 9 May 2015
| title = Automation and anxiety
| newspaper = The Economist
| url = https://www.economist.com/news/special-report/21700758-will-smarter-machines-cause-mass-unemployment-automation-and-anxiety
| access-date = 13 January 2018
| archive-date = 12 January 2018
| archive-url = https://web.archive.org/web/20180112214621/https://www.economist.com/news/special-report/21700758-will-smarter-machines-cause-mass-unemployment-automation-and-anxiety
| url-status = live
}}
* {{cite news
| last = Mahdawi
| first = Arwa
| date = 26 June 2017
| title = What jobs will still be around in 20 years? Read this to prepare your future
| work = The Guardian
| url = https://www.theguardian.com/us-news/2017/jun/26/jobs-future-automation-robots-skills-creative-health
| access-date = 13 January 2018
| archive-date = 14 January 2018
| archive-url = https://web.archive.org/web/20180114021804/https://www.theguardian.com/us-news/2017/jun/26/jobs-future-automation-robots-skills-creative-health
| url-status = live
}}
* {{cite journal |last=Rubin |first=Charles |title=Artificial Intelligence and Human Nature |date=Spring 2003 |journal=The New Atlantis |volume=1 |pages=88–100 |url=http://www.thenewatlantis.com/publications/artificial-intelligence-and-human-nature |archive-url=https://web.archive.org/web/20120611115223/http://www.thenewatlantis.com/publications/artificial-intelligence-and-human-nature |archive-date=11 June 2012 |url-status=dead }}
* {{Cite book
|date=2014
|title=Superintelligence: Paths, Dangers, Strategies |publisher=Oxford University Press |title-link=Superintelligence: Paths, Dangers, Strategies }}
* {{cite web|last=Brooks|first=Rodney|title=artificial intelligence is a tool, not a threat|date=10 November 2014|url=http://www.rethinkrobotics.com/artificial-intelligence-tool-threat/|url-status=dead|archive-url=https://web.archive.org/web/20141112130954/http://www.rethinkrobotics.com/artificial-intelligence-tool-threat/|archive-date=12 November 2014}}
* {{cite web|last=Sainato|first=Michael|title=Stephen Hawking, Elon Musk, and Bill Gates Warn About Artificial Intelligence|date=19 August 2015|url=http://observer.com/2015/08/stephen-hawking-elon-musk-and-bill-gates-warn-about-artificial-intelligence/|url-status=live|archive-url=https://web.archive.org/web/20151030053323/http://observer.com/2015/08/stephen-hawking-elon-musk-and-bill-gates-warn-about-artificial-intelligence/|archive-date=30 October 2015|access-date=30 October 2015|website=Observer}}
* {{cite magazine
| last = Harari
| first = Yuval Noah
| title = Why Technology Favors Tyranny
| date = October 2018
| magazine = The Atlantic
| url = https://www.theatlantic.com/magazine/archive/2018/10/yuval-noah-harari-technology-tyranny/568330
| access-date = 23 September 2021
| archive-date = 25 September 2021
| archive-url = https://web.archive.org/web/20210925221449/https://www.theatlantic.com/magazine/archive/2018/10/yuval-noah-harari-technology-tyranny/568330/
| url-status = live
}}
* {{cite web|last=Robitzski|first=Dan|title=Five experts share what scares them the most about AI|date=5 September 2018|url=https://futurism.com/artificial-intelligence-experts-fear/amp|access-date=8 December 2019|archive-date=8 December 2019|archive-url=https://web.archive.org/web/20191208094101/https://futurism.com/artificial-intelligence-experts-fear/amp|url-status=live}}
* {{cite book|last1=Goffrey|first1=Andrew|editor1-last=Fuller|editor1-first=Matthew|title=Software studies: a lexicon|url=https://archive.org/details/softwarestudiesl00full_007|url-access=limited|date=2008|publisher=MIT Press|location=Cambridge, Mass.|isbn=978-1-4356-4787-9|pages=[https://archive.org/details/softwarestudiesl00full_007/page/n29 15]–20|chapter=Algorithm}}
* {{citation|last1=Lipartito|first1=Kenneth|title=The Narrative and the Algorithm: Genres of Credit Reporting from the Nineteenth Century to Today|date=6 January 2011|doi=10.2139/ssrn.1736283|s2cid=166742927|url=https://mpra.ub.uni-muenchen.de/28142/1/MPRA_paper_28142.pdf|archive-url=https://ghostarchive.org/archive/20221009/https://mpra.ub.uni-muenchen.de/28142/1/MPRA_paper_28142.pdf|archive-date=2022-10-09|url-status=live|type=Unpublished manuscript}}
* {{Cite journal|last1=Goodman|first1=Bryce|last2=Flaxman|first2=Seth|title=EU regulations on algorithmic decision-making and a "right to explanation"|journal=AI Magazine|volume=38|issue=3|pages=50|arxiv=1606.08813|doi=10.1609/aimag.v38i3.2741|year=2017|s2cid=7373959}}
* {{Cite web|author=CNA|title=Commentary: Bad news. Artificial intelligence is biased|date=12 January 2019|url=https://www.channelnewsasia.com/news/commentary/artificial-intelligence-big-data-bias-hiring-loans-key-challenge-11097374|access-date=19 June 2020|website=CNA|language=en|archive-date=12 January 2019|archive-url=https://web.archive.org/web/20190112104421/https://www.channelnewsasia.com/news/commentary/artificial-intelligence-big-data-bias-hiring-loans-key-challenge-11097374|url-status=live}}
* {{Cite web|last1=Larson|first1=Jeff|last2=Angwin|first2=Julia|date=23 May 2016|title=How We Analyzed the COMPAS Recidivism Algorithm|url=https://www.propublica.org/article/how-we-analyzed-the-compas-recidivism-algorithm|access-date=19 June 2020|website=ProPublica|language=en|archive-date=29 April 2019|archive-url=https://web.archive.org/web/20190429190950/https://www.propublica.org/article/how-we-analyzed-the-compas-recidivism-algorithm|url-status=live}}
* {{cite journal|last1=Müller|first1=Vincent C.|last2=Bostrom|first2=Nick|year=2014|title=Future Progress in Artificial Intelligence: A Poll Among Experts|url=http://www.sophia.de/pdf/2014_PT-AI_polls.pdf|url-status=live|journal=AI Matters|volume=1|issue=1|pages=9–11|doi=10.1145/2639475.2639478|s2cid=8510016|archive-url=https://web.archive.org/web/20160115114604/http://www.sophia.de/pdf/2014_PT-AI_polls.pdf|archive-date=15 January 2016}}
* {{Cite news|last1 = Cellan-Jones|first1 = Rory|title = Stephen Hawking warns artificial intelligence could end mankind|date = 2 December 2014|url = https://www.bbc.com/news/technology-30290540|access-date = 30 October 2015|url-status = live|archive-url = https://web.archive.org/web/20151030054329/http://www.bbc.com/news/technology-30290540|archive-date = 30 October 2015|work = BBC News}}
* {{cite news|last1=Rawlinson|first1=Kevin|date=29 January 2015|title=Microsoft's Bill Gates insists AI is a threat|work=BBC News|url=https://www.bbc.co.uk/news/31047780|url-status=live|access-date=30 January 2015|archive-url=https://web.archive.org/web/20150129183607/http://www.bbc.co.uk/news/31047780|archive-date=29 January 2015}}
* {{Cite news|last=Holley|first=Peter|date=28 January 2015|title=Bill Gates on dangers of artificial intelligence: 'I don't understand why some people are not concerned'|newspaper=The Washington Post|url=https://www.washingtonpost.com/news/the-switch/wp/2015/01/28/bill-gates-on-dangers-of-artificial-intelligence-dont-understand-why-some-people-are-not-concerned/|url-status=live|access-date=30 October 2015|archive-url=https://web.archive.org/web/20151030054330/https://www.washingtonpost.com/news/the-switch/wp/2015/01/28/bill-gates-on-dangers-of-artificial-intelligence-dont-understand-why-some-people-are-not-concerned/|archive-date=30 October 2015|issn=0190-8286}}
* {{Cite news|last=Gibbs|first=Samuel|date=27 October 2014|title=Elon Musk: artificial intelligence is our biggest existential threat|work=The Guardian|url=https://www.theguardian.com/technology/2014/oct/27/elon-musk-artificial-intelligence-ai-biggest-existential-threat|url-status=live|access-date=30 October 2015|archive-url=https://web.archive.org/web/20151030054330/http://www.theguardian.com/technology/2014/oct/27/elon-musk-artificial-intelligence-ai-biggest-existential-threat|archive-date=30 October 2015}}
* {{cite web|last=Bostrom|first=Nick|date=2015|title=What happens when our computers get smarter than we are?|url=https://www.ted.com/talks/nick_bostrom_what_happens_when_our_computers_get_smarter_than_we_are/transcript|publisher=[[Software:TED (conference)|TED (conference)]]|access-date=30 January 2020|archive-date=25 July 2020|archive-url=https://web.archive.org/web/20200725005719/https://www.ted.com/talks/nick_bostrom_what_happens_when_our_computers_get_smarter_than_we_are/transcript|url-status=live}}
* {{Cite web |title=Oracle CEO Mark Hurd sees no reason to fear ERP AI |first=Patrick |last=Thibodeau |date=25 March 2019 |url=https://searcherp.techtarget.com/news/252460208/Oracle-CEO-Mark-Hurd-sees-no-reason-to-fear-ERP-AI |access-date=6 May 2019 |website=SearchERP |language=en |archive-date=6 May 2019 |archive-url=https://web.archive.org/web/20190506173749/https://searcherp.techtarget.com/news/252460208/Oracle-CEO-Mark-Hurd-sees-no-reason-to-fear-ERP-AI |url-status=live }}
* {{Cite web|first=Prachi|last=Bhardwaj|date=24 May 2018|title=Mark Zuckerberg responds to Elon Musk's paranoia about AI: 'AI is going to... help keep our communities safe.'|url=https://www.businessinsider.com/mark-zuckerberg-shares-thoughts-elon-musks-ai-2018-5|access-date=6 May 2019|website=Business Insider|archive-date=6 May 2019|archive-url=https://web.archive.org/web/20190506173756/https://www.businessinsider.com/mark-zuckerberg-shares-thoughts-elon-musks-ai-2018-5|url-status=live}}
* {{cite web|first=Edward Moore|last=Geist|title=Is artificial intelligence really an existential threat to humanity?|date=9 August 2015|url=http://thebulletin.org/artificial-intelligence-really-existential-threat-humanity8577|url-status=live|archive-url=https://web.archive.org/web/20151030054330/http://thebulletin.org/artificial-intelligence-really-existential-threat-humanity8577|archive-date=30 October 2015|access-date=30 October 2015|website=Bulletin of the Atomic Scientists}}
* {{cite journal|last=Madrigal|first=Alexis C.|title=The case against killer robots, from a guy actually working on artificial intelligence|date=27 February 2015|url=https://www.hrw.org/report/2012/11/19/losing-humanity/case-against-killer-robots|url-status=live|archive-url=https://web.archive.org/web/20160204175716/http://fusion.net/story/54583/the-case-against-killer-robots-from-a-guy-actually-building-ai/|archive-date=4 February 2016|access-date=31 January 2016|website=Fusion.net}}
* {{cite web|last=Lee|first=Timothy B.|title=Will artificial intelligence destroy humanity? Here are 5 reasons not to worry.|date=22 August 2014|url=https://www.vox.com/2014/8/22/6043635/5-reasons-we-shouldnt-worry-about-super-intelligent-computers-taking|url-status=live|archive-url=https://web.archive.org/web/20151030092203/http://www.vox.com/2014/8/22/6043635/5-reasons-we-shouldnt-worry-about-super-intelligent-computers-taking|archive-date=30 October 2015|access-date=30 October 2015|website=Vox}}
* {{Cite book | ref={{harvid|Law Library of Congress (U.S.). Global Legal Research Directorate|2019}}
|author=Law Library of Congress (U.S.). Global Legal Research Directorate, issuing body.
|title=Regulation of artificial intelligence in selected jurisdictions
|year=2019
|lccn=2019668143 |oclc=1110727808}}
* {{Cite book|last1=Berryhill|first1=Jamie|url=https://oecd-opsi.org/wp-content/uploads/2019/11/AI-Report-Online.pdf|title=Hello, World: Artificial Intelligence and its Use in the Public Sector|last2=Heang|first2=Kévin Kok|last3=Clogher|first3=Rob|last4=McBride|first4=Keegan|publisher=OECD Observatory of Public Sector Innovation|year=2019|location=Paris|access-date=9 August 2020|archive-date=20 December 2019|archive-url=https://web.archive.org/web/20191220021331/https://oecd-opsi.org/wp-content/uploads/2019/11/AI-Report-Online.pdf|url-status=live}}
* {{Cite book|title=Research handbook on the law of artificial intelligence|last1=Barfield |first1=Woodrow |last2=Pagallo |first2=Ugo|year = 2018|isbn=978-1-78643-904-8 |location=Cheltenham, UK |oclc=1039480085}}
* {{Cite journal|last1=Iphofen|first1=Ron|last2=Kritikos|first2=Mihalis|date=3 January 2019|title=Regulating artificial intelligence and robotics: ethics by design in a digital society|journal=Contemporary Social Science|volume=16|issue=2|pages=170–184|doi=10.1080/21582041.2018.1563803|s2cid=59298502|issn=2158-2041}}
* {{Cite journal|last1=Wirtz|first1=Bernd W.|last2=Weyerer|first2=Jan C.|last3=Geyer|first3=Carolin|date=24 July 2018|title=Artificial Intelligence and the Public Sector – Applications and Challenges|journal=International Journal of Public Administration|volume=42|issue=7|pages=596–615|doi=10.1080/01900692.2018.1498103|s2cid=158829602|issn=0190-0692|url=https://zenodo.org/record/3569435|access-date=22 August 2020|archive-date=18 August 2020|archive-url=https://web.archive.org/web/20200818131415/https://zenodo.org/record/3569435|url-status=live}}
* {{Cite journal |last=Buiten |first=Miriam C|date=2019|title=Towards Intelligent Regulation of Artificial Intelligence|journal=European Journal of Risk Regulation|volume=10|issue=1|pages=41–59 |doi=10.1017/err.2019.8|issn=1867-299X |doi-access=free}}
* {{cite book |first=Wendell |last=Wallach |year=2010 |title=Moral Machines |publisher=Oxford University Press}}
* {{Cite web|last=Brown|first=Eileen|date=5 November 2019|website=ZDNet|language=en|url=https://www.zdnet.com/article/half-of-americans-do-not-believe-deepfake-news-could-target-them-online/|title=Half of Americans do not believe deepfake news could target them online|access-date=3 December 2019|archive-url=https://web.archive.org/web/20191106035012/https://www.zdnet.com/article/half-of-americans-do-not-believe-deepfake-news-could-target-them-online/|archive-date=6 November 2019|url-status=live}}
* {{Cite web|last=Frangoul|first=Anmar|date=14 June 2019|website=CNBC|language=en|url=https://www.cnbc.com/2019/06/14/the-business-using-ai-to-change-how-we-think-about-energy-storage.html|title=A Californian business is using A.I. to change the way we think about energy storage|access-date=5 November 2019|archive-date=25 July 2020|archive-url=https://web.archive.org/web/20200725044735/https://www.cnbc.com/2019/06/14/the-business-using-ai-to-change-how-we-think-about-energy-storage.html|url-status=live}}
* {{cite news
| ref={{harvid|Economist|2016}}
| title=The Economist Explains: Why firms are piling into artificial intelligence
| newspaper=[[Finance:The Economist|The Economist]]
| date=31 March 2016
| url=https://www.economist.com/blogs/economist-explains/2016/04/economist-explains
| access-date=19 May 2016
| url-status=live
| archive-url=https://web.archive.org/web/20160508010311/http://www.economist.com/blogs/economist-explains/2016/04/economist-explains
| archive-date=8 May 2016
}}
* {{cite news|last=Lohr|first=Steve|title=The Promise of Artificial Intelligence Unfolds in Small Steps|date=28 February 2016|work={{wipe|The New York Times}}|url=https://www.nytimes.com/2016/02/29/technology/the-promise-of-artificial-intelligence-unfolds-in-small-steps.html|access-date=29 February 2016|url-status=live|archive-url=https://web.archive.org/web/20160229171843/http://www.nytimes.com/2016/02/29/technology/the-promise-of-artificial-intelligence-unfolds-in-small-steps.html|archive-date=29 February 2016}}
* {{cite web|url=https://www.bbc.co.uk/news/business-36837824|title=So you think you chose to read this article?|date=22 July 2016|last=Smith|first=Mark|work=BBC News|url-status=live|archive-url=https://web.archive.org/web/20160725205007/http://www.bbc.co.uk/news/business-36837824|archive-date=25 July 2016}}
* {{cite journal | ref={{harvid|Aletras ''et al.''|2016}}
|first1=N. | last1=Aletras
|first2=D. | last2=Tsarapatsanis
|first3=D. | last3=Preotiuc-Pietro
|first4=V. | last4=Lampos
|title=Predicting judicial decisions of the European Court of Human Rights: a Natural Language Processing perspective
|year=2016
|journal=PeerJ Computer Science |volume=2 |pages=e93
|doi=10.7717/peerj-cs.93 |doi-access=free}}
* {{cite journal |last1=Cadena |first1=Cesar |last2=Carlone |first2=Luca |last3=Carrillo |first3=Henry |last4=Latif |first4=Yasir |last5=Scaramuzza |first5=Davide |last6=Neira |first6=Jose |last7=Reid|first7=Ian|last8=Leonard|first8=John J.|title=Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age|journal=IEEE Transactions on Robotics |date=December 2016|volume=32|issue=6|pages=1309–1332 |doi=10.1109/TRO.2016.2624754 |arxiv=1606.05830 |s2cid=2596787}}
* {{cite journal|last1=Cambria|first1=Erik|last2=White|first2=Bebo|title=Jumping NLP Curves: A Review of Natural Language Processing Research [Review Article]|journal=IEEE Computational Intelligence Magazine|date=May 2014|volume=9|issue=2|pages=48–57|doi=10.1109/MCI.2014.2307227|s2cid=206451986}}
* {{cite news |last1=Vincent |first1=James |title=OpenAI has published the text-generating AI it said was too dangerous to share |url=https://www.theverge.com/2019/11/7/20953040/openai-text-generation-ai-gpt-2-full-model-release-1-5b-parameters |access-date=11 June 2020 |work=The Verge |date=7 November 2019 |language=en |archive-date=11 June 2020 |archive-url=https://web.archive.org/web/20200611054114/https://www.theverge.com/2019/11/7/20953040/openai-text-generation-ai-gpt-2-full-model-release-1-5b-parameters |url-status=live }}
* {{cite journal|last1=Jordan|first1=M. I.|last2=Mitchell|first2=T. M. |title=Machine learning: Trends, perspectives, and prospects |journal=Science |date=16 July 2015 |volume=349 |issue=6245 |pages=255–260 |doi=10.1126/science.aaa8415 |pmid=26185243 |bibcode=2015Sci...349..255J |s2cid=677218}}
* {{cite web
|title=Content: Plug & Pray Film – Artificial Intelligence – Robots -
|author=Maschafilm
|year=2010
|work=plugandpray-film.de
|url-status=live
|url=http://www.plugandpray-film.de/en/content.html
|archive-url=https://web.archive.org/web/20160212040134/http://www.plugandpray-film.de/en/content.html
|archive-date=12 February 2016
}}
* {{cite journal |last=Evans |first=Woody |title=Posthuman Rights: Dimensions of Transhuman Worlds |journal=Teknokultura |volume=12 |issue=2 |date=2015 |doi=10.5209/rev_TK.2015.v12.n2.49072 |doi-access=free}}
* {{cite magazine|last1=Waddell|first1=Kaveh|title=Chatbots Have Entered the Uncanny Valley|url=https://www.theatlantic.com/technology/archive/2017/04/uncanny-valley-digital-assistants/523806/|access-date=24 April 2018|magazine=The Atlantic|date=2018|archive-date=24 April 2018|archive-url=https://web.archive.org/web/20180424202350/https://www.theatlantic.com/technology/archive/2017/04/uncanny-valley-digital-assistants/523806/|url-status=live}}
* {{cite journal|last1=Poria|first1=Soujanya|last2=Cambria|first2=Erik|last3=Bajpai|first3=Rajiv|last4=Hussain|first4=Amir|title=A review of affective computing: From unimodal analysis to multimodal fusion|journal=Information Fusion|date=September 2017|volume=37|pages=98–125|doi=10.1016/j.inffus.2017.02.003|hdl=1893/25490|s2cid=205433041 |url=http://researchrepository.napier.ac.uk/Output/1792429|hdl-access=free|access-date=27 April 2021|archive-date=23 March 2023|archive-url=https://web.archive.org/web/20230323165407/https://www.napier.ac.uk/research-and-innovation/research-search/outputs/a-review-of-affective-computing-from-unimodal-analysis-to-multimodal-fusion|url-status=live}}
* {{cite news|ref={{harvid|BBC|2006}}|date=21 December 2006|title=Robots could demand legal rights|work=BBC News|url=http://news.bbc.co.uk/2/hi/technology/6200005.stm|access-date=3 February 2011|archive-url=https://web.archive.org/web/20191015042628/http://news.bbc.co.uk/2/hi/technology/6200005.stm|archive-date=15 October 2019|url-status=live}}
* {{cite web
| last=Horst
| first=Steven
| title=The Computational Theory of Mind
| website=The Stanford Encyclopedia of Philosophy
| year=2005
| url=http://plato.stanford.edu/entries/computational-mind
| access-date=7 March 2016
| archive-date=6 March 2016
| archive-url=https://web.archive.org/web/20160306083748/http://plato.stanford.edu/entries/computational-mind/
| url-status=live
}}
* {{Cite conference | last = Omohundro|first= Steve|  year = 2008| title= The Nature of Self-Improving Artificial Intelligence| publisher=presented and distributed at the 2007 Singularity Summit, San Francisco, CA.}}
* {{cite news
| last1=Ford
| first1=Martin
| last2=Colvin
| first2=Geoff
| date=6 September 2015
| title=Will robots create more jobs than they destroy?
| work=The Guardian
| url=https://www.theguardian.com/technology/2015/sep/06/will-robots-create-destroy-jobs
| access-date=13 January 2018
| archive-date=16 June 2018
| archive-url=https://web.archive.org/web/20180616204119/https://www.theguardian.com/technology/2015/sep/06/will-robots-create-destroy-jobs
| url-status=live
}}
* {{Cite book
| ref={{harvid|European Commission|2020}}
| url=https://ec.europa.eu/info/sites/info/files/commission-white-paper-artificial-intelligence-feb2020_en.pdf
| title=White Paper: On Artificial Intelligence – A European approach to excellence and trust
| publisher=European Commission
| year=2020
| location=Brussels
| access-date=20 February 2020
| archive-date=20 February 2020
| archive-url=https://web.archive.org/web/20200220173419/https://ec.europa.eu/info/sites/info/files/commission-white-paper-artificial-intelligence-feb2020_en.pdf
| url-status=live
}}
* {{cite book
| first1 = Michael | last1 = Anderson
| first2 = Susan Leigh | last2 = Anderson
| year = 2011
| title = Machine Ethics
| publisher = Cambridge University Press.
}}
* {{cite web |ref={{harvid|AAAI|2014}} |title=Machine Ethics |url=http://www.aaai.org/Library/Symposia/Fall/fs05-06 |work=aaai.org |url-status=dead |archive-url=https://web.archive.org/web/20141129044821/http://www.aaai.org/Library/Symposia/Fall/fs05-06 |archive-date=29 November 2014 }}
* {{cite book|last=Russell|first=Stuart|title=Human Compatible: Artificial Intelligence and the Problem of Control|title-link=Human Compatible|date=8 October 2019|publisher=Viking|isbn=978-0-525-55861-3|location=United States|oclc=1083694322|}}
* {{cite news |ref={{harvid|CNN|2006}} |title=AI set to exceed human brain power |work=CNN |date=9 August 2006 |url=http://www.cnn.com/2006/TECH/science/07/24/ai.bostrom/ |archive-url=https://web.archive.org/web/20080219001624/http://www.cnn.com/2006/TECH/science/07/24/ai.bostrom/ |archive-date=19 February 2008 |url-status=live }}
* {{cite news|date=21 December 2006|title=Robots could demand legal rights|work=BBC News|url=http://news.bbc.co.uk/2/hi/technology/6200005.stm|access-date=3 February 2011|archive-url=https://web.archive.org/web/20191015042628/http://news.bbc.co.uk/2/hi/technology/6200005.stm|archive-date=15 October 2019|url-status=live}}
* {{cite web|ref={{harvid|MIT AIL|2014}}|title=Kismet|publisher=MIT Artificial Intelligence Laboratory, Humanoid Robotics Group|url=http://www.ai.mit.edu/projects/humanoid-robotics-group/kismet/kismet.html|access-date=25 October 2014|archive-url=https://web.archive.org/web/20141017040432/http://www.ai.mit.edu/projects/humanoid-robotics-group/kismet/kismet.html|archive-date=17 October 2014|url-status=live}}
* {{cite journal|last1=Smoliar|first1=Stephen W. |last2=Zhang |first2=HongJiang |title=Content based video indexing and retrieval |journal=IEEE MultiMedia |date=1994 |volume=1 |issue=2 |pages=62–72 |doi=10.1109/93.311653 |s2cid=32710913}}
* {{cite journal |last1=Neumann |first1=Bernd |last2=Möller |first2=Ralf |title=On scene interpretation with description logics |journal=Image and Vision Computing |date=January 2008 |volume=26 |issue=1 |pages=82–101 |doi=10.1016/j.imavis.2007.08.013|s2cid=10767011 }}
* {{cite journal |last1=Kuperman |first1=G. J.|last2=Reichley|first2=R. M.|last3=Bailey|first3=T. C.|title=Using Commercial Knowledge Bases for Clinical Decision Support: Opportunities, Hurdles, and Recommendations |journal=Journal of the American Medical Informatics Association|date=1 July 2006|volume=13|issue=4|pages=369–371 |doi=10.1197/jamia.M2055 |pmid=16622160|pmc=1513681}}
* {{cite journal|last1=McGarry|first1=Ken|s2cid=14987656|title=A survey of interestingness measures for knowledge discovery|journal=The Knowledge Engineering Review|date=1 December 2005 |volume=20|issue=1|pages=39–61|doi=10.1017/S0269888905000408}}
* {{cite conference |title=Automatic annotation and semantic retrieval of video sequences using multimedia ontologies |last1=Bertini |first1=M |last2=Del Bimbo |first2=A |last3=Torniai |first3=C |date=2006 |publisher=ACM |book-title=MM '06 Proceedings of the 14th ACM international conference on Multimedia |pages=679–682 |location=Santa Barbara |conference=14th ACM international conference on Multimedia}}
* {{cite book
| last = Kahneman
| first = Daniel
| title = Thinking, Fast and Slow
| date = 25 October 2011
| publisher = Macmillan
| isbn = 978-1-4299-6935-2
| url = https://books.google.com/books?id=ZuKTvERuPG8C
| access-date = 8 April 2012
| archive-date = 15 March 2023
| archive-url = https://web.archive.org/web/20230315191803/https://books.google.com/books?id=ZuKTvERuPG8C
| url-status = live
}}
* {{Citation | last = Turing | first = Alan |  year=1948 | chapter=Machine Intelligence | title = The Essential Turing: The ideas that gave birth to the computer age | editor=Copeland, B. Jack | isbn = 978-0-19-825080-7 | publisher = Oxford University Press | location = Oxford | page = 412}}
* {{cite book
| title = The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World
| date = 22 September 2015
| publisher = Basic Books
| isbn = 978-0465065707
}}
* {{Citation | first = Marvin | last = Minsky | title = The Society of Mind | publisher = Simon and Schuster | year = 1986 |  title-link = The Society of Mind }}
* {{Citation | last=Pinker | first=Steven |  title=The Language Instinct | date=4 September 2007 | orig-year=1994 | publisher=Harper | series = Perennial Modern Classics | isbn=978-0-06-133646-1| title-link=The Language Instinct }}
* {{cite journal |last=Chalmers |first=David |title=Facing up to the problem of consciousness |year=1995 |journal=[[Philosophy:Journal of Consciousness Studies|Journal of Consciousness Studies]] |volume=2 |issue=3 |pages=200–219 |url=http://www.imprint.co.uk/chalmers.html |access-date=11 October 2018 |archive-date=8 March 2005 |archive-url=https://web.archive.org/web/20050308163649/http://www.imprint.co.uk/chalmers.html |url-status=live }}
* {{cite magazine|last1=Roberts|first1=Jacob|title=Thinking Machines: The Search for Artificial Intelligence|magazine=Distillations|volume=2|issue=2|pages=14–23|date=2016|url=https://www.sciencehistory.org/distillations/magazine/thinking-machines-the-search-for-artificial-intelligence|access-date=20 March 2018|archive-url=https://web.archive.org/web/20180819152455/https://www.sciencehistory.org/distillations/magazine/thinking-machines-the-search-for-artificial-intelligence|archive-date=19 August 2018|url-status=dead}}
* {{cite book
| last1=Pennachin|first1=C. |last2=Goertzel|first2=B.
| chapter=Contemporary Approaches to Artificial General Intelligence
| title=Artificial General Intelligence
| date=2007 |doi=10.1007/978-3-540-68677-4_1 |publisher=Springer|location=Berlin, Heidelberg |series=Cognitive Technologies |isbn=978-3-540-23733-4}}
* {{cite web|ref={{harvid|McKinsey|2018}}|title=Ask the AI experts: What's driving today's progress in AI?|url=https://www.mckinsey.com/business-functions/mckinsey-analytics/our-insights/ask-the-ai-experts-whats-driving-todays-progress-in-ai|website=McKinsey & Company|access-date=13 April 2018|archive-date=13 April 2018|archive-url=https://web.archive.org/web/20180413190018/https://www.mckinsey.com/business-functions/mckinsey-analytics/our-insights/ask-the-ai-experts-whats-driving-todays-progress-in-ai|url-status=live}}
* {{cite journal
| ref = {{harvid|MIT Sloan Management Review|2018}}
| title = Reshaping Business With Artificial Intelligence
| journal = MIT Sloan Management Review
| date = 6 September 2017
| url = https://sloanreview.mit.edu/projects/reshaping-business-with-artificial-intelligence/
| access-date = 2 May 2018
| archive-date = 19 May 2018
| archive-url = https://web.archive.org/web/20180519171905/https://sloanreview.mit.edu/projects/reshaping-business-with-artificial-intelligence/
| url-status = live
| last1 = Ransbotham
| first1 = Sam
| last2 = Kiron
| first2 = David
| last3 = Gerbert
| first3 = Philipp
| last4 = Reeves
| first4 = Martin
}}
* {{cite web|last1=Lorica|first1=Ben|title=The state of AI adoption|date=18 December 2017|url=https://www.oreilly.com/ideas/the-state-of-ai-adoption|website=O'Reilly Media|access-date=2 May 2018|archive-date=2 May 2018|archive-url=https://web.archive.org/web/20180502140700/https://www.oreilly.com/ideas/the-state-of-ai-adoption|url-status=live}}
* {{cite web
| ref = {{Harvid|Google|2016}}
| title = AlphaGo – Google DeepMind
| url = https://wildoftech.com/alphago-google-deepmind
| url-status = live
| archive-url = https://web.archive.org/web/20160310191926/https://wildoftech.com/alphago-google-deepmind/
| archive-date = 20 October 2021
}}
* {{cite journal
|last1=Asada |first1=M. |last2=Hosoda |first2=K. |last3=Kuniyoshi |first3=Y. |last4=Ishiguro |first4=H. |last5=Inui |first5=T. |last6=Yoshikawa |first6=Y. |last7=Ogino |first7=M. |last8=Yoshida |first8=C. |year=2009 |title=Cognitive developmental robotics: a survey |journal=IEEE Transactions on Autonomous Mental Development |volume=1 |issue=1 |pages=12–34 |doi=10.1109/tamd.2009.2021702 |s2cid=10168773 }}
* {{cite book | first = David | last = Berlinski | year = 2000 | title = The Advent of the Algorithm | publisher = Harcourt Books |  isbn = 978-0-15-601391-8 | oclc = 46890682 | url = https://archive.org/details/adventofalgorith0000berl | access-date = 22 August 2020 | archive-date = 26 July 2020 | archive-url = https://web.archive.org/web/20200726215744/https://archive.org/details/adventofalgorith0000berl | url-status = live }}
* {{cite journal | last=Brooks | first=Rodney |  year=1990 | title=Elephants Don't Play Chess | journal=Robotics and Autonomous Systems | volume=6 | issue=1–2 | pages=3–15 | doi=10.1016/S0921-8890(05)80025-9 | url=http://people.csail.mit.edu/brooks/papers/elephants.pdf | archive-url=https://web.archive.org/web/20070809020912/http://people.csail.mit.edu/brooks/papers/elephants.pdf | archive-date=9 August 2007 | url-status=live | citeseerx=10.1.1.588.7539 }}
* {{cite news
|last=Butler
|first=Samuel
|date=13 June 1863
|title=Darwin among the Machines
|work=The Press
|location=Christchurch, New Zealand
|department=Letters to the Editor
|url=https://nzetc.victoria.ac.nz/tm/scholarly/tei-ButFir-t1-g1-t1-g1-t4-body.html
|access-date=16 October 2014
|via=Victoria University of Wellington
|archive-date=19 September 2008
|archive-url=https://web.archive.org/web/20080919172551/http://www.nzetc.org/tm/scholarly/tei-ButFir-t1-g1-t1-g1-t4-body.html
|url-status=live
}}
* {{cite news
 |url = https://www.bloomberg.com/news/articles/2015-12-08/why-2015-was-a-breakthrough-year-in-artificial-intelligence
 |title = Why 2015 Was a Breakthrough Year in Artificial Intelligence
 |last = Clark
 |first = Jack
 |website = Bloomberg.com
 |year = 2015b
 |url-access = subscription
 |access-date = 23 November 2016
 |url-status = live
 |archive-url = https://web.archive.org/web/20161123053855/https://www.bloomberg.com/news/articles/2015-12-08/why-2015-was-a-breakthrough-year-in-artificial-intelligence
 |archive-date = 23 November 2016
 }}
* {{cite book
| year=1991
| title=Consciousness Explained
| publisher=The Penguin Press
| isbn= 978-0-7139-9037-9
| title-link=Consciousness Explained }}
* {{cite book
| year = 1972
| title = What Computers Can't Do
| publisher = MIT Press | location = New York
| isbn = 978-0-06-011082-6
| title-link=What Computers Can't Do }}
* {{cite book
 |last1 = Dreyfus
 |first1 = Hubert
 |last2 = Dreyfus
 |first2 = Stuart
 |year = 1986
 |title = Mind over Machine: The Power of Human Intuition and Expertise in the Era of the Computer
 |publisher = Blackwell
 |location = Oxford, UK
 |isbn = 978-0-02-908060-3
 |url = https://archive.org/details/mindovermachinep00drey
 |access-date = 22 August 2020
 |archive-date = 26 July 2020
 |archive-url = https://web.archive.org/web/20200726131414/https://archive.org/details/mindovermachinep00drey
 |url-status = live
 }}
* {{cite book
|last=Dyson
|first=George
|year=1998
|title=Darwin among the Machines
|publisher=Allan Lane Science
|isbn=978-0-7382-0030-9
|url=https://archive.org/details/darwinamongmachi00dyso
|access-date=22 August 2020
|archive-date=26 July 2020
|archive-url=https://web.archive.org/web/20200726131443/https://archive.org/details/darwinamongmachi00dyso
|url-status=live
}}
* {{cite book
|last=Edelson
|first=Edward
|year=1991
|title=The Nervous System
|url=https://archive.org/details/nervoussystem0000edel
|url-access=registration
|publisher=Chelsea House<!--so Worldcat, originally here Remmel Nunn-->
|location=New York
|isbn=978-0-7910-0464-7
|access-date=18 November 2019
|archive-date=26 July 2020
|archive-url=https://web.archive.org/web/20200726131758/https://archive.org/details/nervoussystem0000edel
|url-status=live
}}
* {{cite book
| last=Fearn
| first = Nicholas
| year =2007
| title= The Latest Answers to the Oldest Questions: A Philosophical Adventure with the World's Greatest Thinkers
| publisher = Grove Press
| location=New York |isbn=978-0-8021-1839-4
}}
* {{cite book
| year = 1985
| title = Artificial Intelligence: The Very Idea
| publisher=MIT Press| location= Cambridge, Mass.
| isbn=978-0-262-08153-5
}}
* {{cite book
| last2=Blakeslee | first2=Sandra
| year=2005
| title=On Intelligence
| publisher=Owl Books | location=New York
| isbn=978-0-8050-7853-4
| title-link=On Intelligence }}
* {{cite news
| last=Henderson
| first=Mark
| date=24 April 2007
| title=Human rights for robots? We're getting carried away
| url=http://www.thetimes.co.uk/tto/technology/article1966391.ece
| work=The Times Online
| location=London
| access-date=31 May 2014
| archive-date=31 May 2014
| archive-url=https://web.archive.org/web/20140531104850/http://www.thetimes.co.uk/tto/technology/article1966391.ece
| url-status=live
}}
* {{cite book
| last2=Slovic | first2= D.
| year=1982
| title=Judgment under uncertainty: Heuristics and biases
| journal=Science | volume=185 | issue=4157 | pages=1124–1131 | publisher=Cambridge University Press | location=New York
| isbn=978-0-521-28414-1
| pmid=17835457 | doi=10.1126/science.185.4157.1124 | bibcode=1974Sci...185.1124T | s2cid=143452957 }}
* {{cite magazine
|last=Katz
|first=Yarden
|date=1 November 2012
|title=Noam Chomsky on Where Artificial Intelligence Went Wrong
|magazine=The Atlantic
|url=https://www.theatlantic.com/technology/archive/2012/11/noam-chomsky-on-where-artificial-intelligence-went-wrong/261637/?single_page=true
|access-date=26 October 2014
|archive-date=28 February 2019
|archive-url=https://web.archive.org/web/20190228154403/https://www.theatlantic.com/technology/archive/2012/11/noam-chomsky-on-where-artificial-intelligence-went-wrong/261637/?single_page=true
|url-status=live
}}
* {{cite book
| year=2005
| title=The Singularity is Near
| publisher=Penguin Books
| isbn=978-0-670-03384-3
| title-link=The Singularity is Near }}
* {{Cite journal
| last1=Langley |first1=Pat |year=2011
| title=The changing science of machine learning
| journal=Machine Learning
| volume=82 |issue=3 |pages=275–279
| doi=10.1007/s10994-011-5242-y
|doi-access=free
}}
* {{cite arXiv
| last1=Legg |first1=Shane |last2=Hutter |first2=Marcus |date=15 June 2007
| title=A Collection of Definitions of Intelligence
|class=cs.AI |eprint=0706.3639
}}
* {{cite book
| last2=Guha | first2=R. V.
| year = 1989
| title = Building Large Knowledge-Based Systems
| publisher = Addison-Wesley
| isbn=978-0-201-51752-1
}}
* {{Cite book
| year=1973
| contribution= Artificial Intelligence: A General Survey
| title=Artificial Intelligence: a paper symposium
| publisher=Science Research Council
}}
* {{cite journal
| last1 = Lombardo| first1 = P| last2 = Boehm | first2 = I| last3 = Nairz| first3 = K
| year = 2020
| title = RadioComics – Santa Claus and the future of radiology
| journal = Eur J Radiol | volume = 122 | issue = 1 | pages=108771
| pmid = 31835078
| doi = 10.1016/j.ejrad.2019.108771|doi-access=free}}
* {{cite journal
| last1=Lungarella |first1=M. |last2=Metta |first2=G. |last3=Pfeifer |first3=R. |last4=Sandini |first4=G. |year=2003
| title=Developmental robotics: a survey
| journal=Connection Science |volume=15 | issue=4 |pages=151–190 |citeseerx=10.1.1.83.7615 | doi=10.1080/09540090310001655110
| s2cid=1452734 }}
* {{cite web
 |last=Maker
 |first=Meg Houston
 |year=2006
 |title=AI@50: AI Past, Present, Future
 |location=Dartmouth College
 |url=http://www.engagingexperience.com/2006/07/ai50_ai_past_pr.html
 |archive-url=https://web.archive.org/web/20070103222615/http://www.engagingexperience.com/2006/07/ai50_ai_past_pr.html
 |url-status=dead
 |archive-date=3 January 2007
 |access-date=16 October 2008
 }}
* {{cite web | last1 = McCarthy | first1 = John |  last2 = Minsky | first2 = Marvin | author-link2 = Marvin Minsky | last3 = Rochester | first3 = Nathan | author-link3 = Nathan Rochester | last4 = Shannon | first4 = Claude | author-link4 = Claude Shannon | year = 1955 | title = A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence | url = http://www-formal.stanford.edu/jmc/history/dartmouth/dartmouth.html | access-date = 30 August 2007 | archive-url = https://web.archive.org/web/20070826230310/http://www-formal.stanford.edu/jmc/history/dartmouth/dartmouth.html | archive-date = 26 August 2007 | url-status = dead }}
* {{cite book
 |last = Minsky
 |first = Marvin
 |year = 1967
 |title = Computation: Finite and Infinite Machines
 |url = https://archive.org/details/computationfinit0000mins
 |url-access = registration
 |publisher = Prentice-Hall
 |location = Englewood Cliffs, N.J.
 |isbn = 978-0-13-165449-5
 |access-date = 18 November 2019
 |archive-date = 26 July 2020
 |archive-url = https://web.archive.org/web/20200726131743/https://archive.org/details/computationfinit0000mins
 |url-status = live
 }}
* {{cite book
|last=Moravec
|first=Hans
|year=1988
|title=Mind Children
|url=https://archive.org/details/mindchildrenfutu00mora
|url-access=registration
|publisher=Harvard University Press
|isbn=978-0-674-57616-2
|access-date=18 November 2019
|archive-date=26 July 2020
|archive-url=https://web.archive.org/web/20200726131644/https://archive.org/details/mindchildrenfutu00mora
|url-status=live
}}
* {{cite book |ref={{harvid|NRC|1999}}
| year=1999
| chapter=Developments in Artificial Intelligence
| title=Funding a Revolution: Government Support for Computing Research
| publisher=National Academy Press
}}
* {{cite journal |last1=Newell |first1=Allen |last2=Simon |first2=H. A. |author-link2=Herbert A. Simon |year=1976 |title=Computer Science as Empirical Inquiry: Symbols and Search |journal=Communications of the ACM |volume=19 |issue=3 |doi=10.1145/360018.360022 |pages=113–126 |doi-access=free }}.
* {{Cite journal
 |last = Nilsson
 |first = Nils
 |year = 1983
 |title = Artificial Intelligence Prepares for 2001
 |journal = AI Magazine
 |volume = 1
 |number = 1
 |url = https://ai.stanford.edu/~nilsson/OnlinePubs-Nils/General%20Essays/AIMag04-04-002.pdf
 |access-date = 22 August 2020
 |archive-date = 17 August 2020
 |archive-url = https://web.archive.org/web/20200817194457/http://ai.stanford.edu/~nilsson/OnlinePubs-Nils/General%20Essays/AIMag04-04-002.pdf
 |url-status = live
 }} Presidential Address to the [[Organization:Association for the Advancement of Artificial Intelligence|Association for the Advancement of Artificial Intelligence]].
* {{cite journal | last=Oudeyer | first=P-Y. | year=2010 | title=On the impact of robotics in behavioral and cognitive sciences: from insect navigation to human cognitive development | journal=IEEE Transactions on Autonomous Mental Development | volume=2 | issue=1 | pages=2–16 | url=http://www.pyoudeyer.com/IEEETAMDOudeyer10.pdf | doi=10.1109/tamd.2009.2039057 | s2cid=6362217 | access-date=4 June 2013 | archive-date=3 October 2018 | archive-url=https://web.archive.org/web/20181003202543/http://www.pyoudeyer.com/IEEETAMDOudeyer10.pdf | url-status=live }}
* {{cite magazine
|last=Schank |first=Roger C.
|title=Where's the AI
|magazine=AI magazine |volume=12 |issue=4
|year=1991
}}
* {{Cite journal
 |last=Searle
 |first=John
 |year=1980
 |title=Minds, Brains and Programs
 |journal=Behavioral and Brain Sciences
 |volume=3
 |issue=3
 |pages=417–457
 |doi=10.1017/S0140525X00005756
 |s2cid=55303721
 |url=http://cogprints.org/7150/1/10.1.1.83.5248.pdf
 |access-date=22 August 2020
 |archive-date=17 March 2019
 |archive-url=https://web.archive.org/web/20190317230215/http://cogprints.org/7150/1/10.1.1.83.5248.pdf
 |url-status=live
 }}
* {{cite book
 |last = Searle
 |first = John
 |year = 1999
 |title = Mind, language and society
 |publisher = Basic Books
 |location = New York
 |isbn = 978-0-465-04521-1
 |oclc = 231867665
 |url = https://archive.org/details/mindlanguagesoci00sear
 |access-date = 22 August 2020
 |archive-date = 26 July 2020
 |archive-url = https://web.archive.org/web/20200726220615/https://archive.org/details/mindlanguagesoci00sear
 |url-status = live
 }}
* {{cite book
| last = Simon
| first = H. A.
| year = 1965
| title = The Shape of Automation for Men and Management
| url = https://archive.org/details/shapeofautomatio00simo
| url-access = registration
| publisher = Harper & Row
| location = New York
| access-date = 18 November 2019
| archive-date = 26 July 2020
| archive-url = https://web.archive.org/web/20200726131655/https://archive.org/details/shapeofautomatio00simo
| url-status = live
}}
* {{cite conference
| last=Solomonoff
| first=Ray
| year=1956
| title=An Inductive Inference Machine
| conference=Dartmouth Summer Research Conference on Artificial Intelligence
| url=http://world.std.com/~rjs/indinf56.pdf
| via=std.com, pdf scanned copy of the original
| access-date=22 March 2011
| archive-date=26 April 2011
| archive-url=https://web.archive.org/web/20110426161749/http://world.std.com/~rjs/indinf56.pdf
| url-status=live
}} Later published as<br />{{cite book | ref = none
| last=Solomonoff |first=Ray |year=1957 |pages=56–62
| chapter=An Inductive Inference Machine
| title=IRE Convention Record |volume=Section on Information Theory, part 2
}}
* {{cite web
 |last         = Spadafora
 |first        = Anthony
 |date         = 21 October 2016
 |title        = Stephen Hawking believes AI could be mankind's last accomplishment
 |website      = BetaNews
 |url          = https://betanews.com/2016/10/21/artificial-intelligence-stephen-hawking/
 |url-status   = live
 |archive-url  = https://web.archive.org/web/20170828183930/https://betanews.com/2016/10/21/artificial-intelligence-stephen-hawking/
 |archive-date = 28 August 2017
}}
* {{cite conference
| last1=Tao |first1=Jianhua |first2=Tieniu |last2=Tan |series=Lecture Notes in Computer Science |year=2005
| conference=Affective Computing: A Review
| book-title=Affective Computing and Intelligent Interaction |volume=LNCS 3784 |pages=981–995
 |publisher=Springer |doi=10.1007/11573548
|isbn=978-3-540-29621-8 }}
* {{cite journal
| last=Tecuci |first=Gheorghe |date=March–April 2012
| title=Artificial Intelligence
| journal=Wiley Interdisciplinary Reviews: Computational Statistics
| volume=4 |issue=2 |pages=168–180 |doi=10.1002/wics.200
|s2cid=196141190 }}
* {{cite book
|last=Thro
|first=Ellen
|year=1993
|title=Robotics: The Marriage of Computers and Machines
|location=New York
|publisher=Facts on File
|isbn=978-0-8160-2628-9
|url=https://archive.org/details/isbn_9780816026289
|access-date=22 August 2020
|archive-date=26 July 2020
|archive-url=https://web.archive.org/web/20200726131505/https://archive.org/details/isbn_9780816026289
|url-status=live
}}
* {{Turing 1950}}.
* {{cite book
|ref={{harvid|UNESCO|2021}}
|title=UNESCO Science Report: the Race Against Time for Smarter Development.
|date=11 June 2021
|publisher=UNESCO
|location=Paris
|isbn=978-92-3-100450-6
|url=https://unesdoc.unesco.org/ark:/48223/pf0000377433/PDF/377433eng.pdf.multi
|access-date=18 September 2021
|archive-date=18 June 2022
|archive-url=https://web.archive.org/web/20220618233752/https://unesdoc.unesco.org/ark:/48223/pf0000377433/PDF/377433eng.pdf.multi
|url-status=live
}}
* {{cite journal
| last=Vinge
| first=Vernor
| year=1993
| title=The Coming Technological Singularity: How to Survive in the Post-Human Era
| journal=Vision 21: Interdisciplinary Science and Engineering in the Era of Cyberspace
| pages=11
| url=http://www-rohan.sdsu.edu/faculty/vinge/misc/singularity.html
| bibcode=1993vise.nasa...11V
| access-date=14 November 2011
| archive-date=1 January 2007
| archive-url=https://web.archive.org/web/20070101133646/http://www-rohan.sdsu.edu/faculty/vinge/misc/singularity.html
| url-status=dead
}}
* {{cite book
| last1=Wason
| first1=P. C.
| last2=Shapiro
| first2=D.
| editor=Foss, B. M.
| year=1966
| title=New horizons in psychology
| chapter-url=https://archive.org/details/newhorizonsinpsy0000foss
| chapter-url-access=registration
| location=Harmondsworth
| publisher=Penguin
| chapter=Reasoning
| access-date=18 November 2019
| archive-date=26 July 2020
| archive-url=https://web.archive.org/web/20200726131518/https://archive.org/details/newhorizonsinpsy0000foss
| url-status=live
}}
* {{cite journal |last1=Weng |first1=J. |last2=McClelland |last3=Pentland |first3=A. |last4=Sporns |first4=O. |last5=Stockman |first5=I. |last6=Sur |first6=M. |last7=Thelen |first7=E. |year=2001 |url=http://www.cse.msu.edu/dl/SciencePaper.pdf |via=msu.edu |doi=10.1126/science.291.5504.599 |pmid=11229402 |title=Autonomous mental development by robots and animals |journal=Science |volume=291 |issue=5504 |pages=599–600 |s2cid=54131797 |access-date=4 June 2013 |archive-date=4 September 2013 |archive-url=https://web.archive.org/web/20130904235242/http://www.cse.msu.edu/dl/SciencePaper.pdf |url-status=live }}
{{Refend}}

== Further reading ==
{{refbegin|30em}}

* Autor, David H., "Why Are There Still So Many Jobs? The History and Future of Workplace Automation" (2015) 29(3) ''Journal of Economic Perspectives'' 3.
* [[Biography:Margaret Boden|Boden, Margaret]], ''Mind As Machine'', [[Organization:Oxford University Press|Oxford University Press]], 2006.
* Cukier, Kenneth, "Ready for Robots? How to Think about the Future of AI", ''Foreign Affairs'', vol. 98, no. 4 (July/August 2019), pp.&nbsp;192–98. [[Biography:George Dyson (science historian)|George Dyson]], historian of computing, writes (in what might be called "Dyson's Law") that "Any system simple enough to be understandable will not be complicated enough to behave intelligently, while any system complicated enough to behave intelligently will be too complicated to understand." (p.&nbsp;197.) Computer scientist [[Biography:Alex Pentland|Alex Pentland]] writes: "Current [[Machine learning|AI machine-learning]] [[Algorithm|algorithm]]s are, at their core, dead simple stupid. They work, but they work by brute force." (p.&nbsp;198.)
* [[Biography:Pedro Domingos|Domingos, Pedro]], "Our Digital Doubles: AI will serve our species, not control it", ''Scientific American'', vol. 319, no. 3 (September 2018), pp.&nbsp;88–93.
* Johnston, John (2008) ''The Allure of Machinic Life: Cybernetics, Artificial Life, and the New AI'', MIT Press.
* {{cite journal |last1=Jumper |first1=John |last2=Evans |first2=Richard |last3=Pritzel |first3=Alexander |last4=Green |first4=Tim |last5=Figurnov |first5=Michael |last6=Ronneberger |first6=Olaf |last7=Tunyasuvunakool |first7=Kathryn |last8=Bates |first8=Russ |last9=Žídek |first9=Augustin |last10=Potapenko |first10=Anna |last11=Bridgland |first11=Alex |last12=Meyer |first12=Clemens |last13=Kohl |first13=Simon A. A. |last14=Ballard |first14=Andrew J. |last15=Cowie |first15=Andrew |last16=Romera-Paredes |first16=Bernardino |last17=Nikolov |first17=Stanislav |last18=Jain |first18=Rishub |last19=Adler |first19=Jonas |last20=Back |first20=Trevor |last21=Petersen |first21=Stig |last22=Reiman |first22=David |last23=Clancy |first23=Ellen |last24=Zielinski |first24=Michal |last25=Steinegger |first25=Martin |last26=Pacholska |first26=Michalina |last27=Berghammer |first27=Tamas |last28=Bodenstein |first28=Sebastian |last29=Silver |first29=David |last30=Vinyals |first30=Oriol |last31=Senior |first31=Andrew W. |last32=Kavukcuoglu |first32=Koray |last33=Kohli |first33=Pushmeet |last34=Hassabis |first34=Demis | display-authors=3 | title=Highly accurate protein structure prediction with AlphaFold |journal=Nature |date=26 August 2021 |volume=596 |issue=7873 |pages=583–589 |doi=10.1038/s41586-021-03819-2 |url=https://www.nature.com/articles/s41586-021-03819-2}}
* {{cite journal |last1=LeCun |first1=Yann |last2=Bengio |first2=Yoshua |last3=Hinton |first3=Geoffrey |title=Deep learning |journal=Nature |date=28 May 2015 |volume=521 |issue=7553 |pages=436–444 |doi=10.1038/nature14539 |url=https://www.nature.com/articles/nature14539}}
* [[Biography:Gary Marcus|Gary Marcus]], "Artificial Confidence: Even the newest, buzziest systems of artificial general intelligence are stymmied by the same old problems", ''Scientific American'', vol. 327, no. 4 (October 2022), pp.&nbsp;42–45. 
* {{cite book |last1=Mitchell |first1=Melanie |title=Artificial intelligence: a guide for thinking humans |date=2019 |publisher=Farrar, Straus and Giroux |location=New York |isbn=9780374257835}}
* {{cite journal |last1=Mnih |first1=Volodymyr |last2=Kavukcuoglu |first2=Koray |last3=Silver |first3=David |last4=Rusu |first4=Andrei A. |last5=Veness |first5=Joel |last6=Bellemare |first6=Marc G. |last7=Graves |first7=Alex |last8=Riedmiller |first8=Martin |last9=Fidjeland |first9=Andreas K. |last10=Ostrovski |first10=Georg |last11=Petersen |first11=Stig |last12=Beattie |first12=Charles |last13=Sadik |first13=Amir |last14=Antonoglou |first14=Ioannis |last15=King |first15=Helen |last16=Kumaran |first16=Dharshan |last17=Wierstra |first17=Daan |last18=Legg |first18=Shane |last19=Hassabis |first19=Demis |display-authors=3 |title=Human-level control through deep reinforcement learning |journal=Nature |date=26 February 2015 |volume=518 |issue=7540 |pages=529–533 |doi=10.1038/nature14236 |url=https://www.nature.com/articles/nature14236/}} Introduced DQN, which produced human-level performance on some Atari games.
* {{cite journal | last1 = Serenko | first1 = Alexander | author2 = Michael Dohan | year = 2011 | title = Comparing the expert survey and citation impact journal ranking methods: Example from the field of Artificial Intelligence | url = http://www.aserenko.com/papers/JOI_AI_Journal_Ranking_Serenko.pdf | journal = Journal of Informetrics | volume = 5 | issue = 4 | pages = 629–49 | doi = 10.1016/j.joi.2011.06.002 | access-date = 12 September 2013 | archive-date = 4 October 2013 | archive-url = https://web.archive.org/web/20131004212839/http://www.aserenko.com/papers/JOI_AI_Journal_Ranking_Serenko.pdf | url-status = live }}
* {{cite journal |last1=Silver |first1=David |last2=Huang |first2=Aja |last3=Maddison |first3=Chris J. |last4=Guez |first4=Arthur |last5=Sifre |first5=Laurent |last6=van den Driessche |first6=George |last7=Schrittwieser |first7=Julian |last8=Antonoglou |first8=Ioannis |last9=Panneershelvam |first9=Veda |last10=Lanctot |first10=Marc |last11=Dieleman |first11=Sander |last12=Grewe |first12=Dominik |last13=Nham |first13=John |last14=Kalchbrenner |first14=Nal |last15=Sutskever |first15=Ilya |last16=Lillicrap |first16=Timothy |last17=Leach |first17=Madeleine |last18=Kavukcuoglu |first18=Koray |last19=Graepel |first19=Thore |last20=Hassabis |first20=Demis | display-authors=3 | title=Mastering the game of Go with deep neural networks and tree search |journal=Nature |date=28 January 2016 |volume=529 |issue=7587 |pages=484–489 |doi=10.1038/nature16961 |url=https://www.nature.com/articles/nature16961}}
* Ashish Vaswani, Noam Shazeer, Niki Parmar et al. "Attention is all you need." Advances in neural information processing systems 30 (2017). Seminal paper on [[Transformer (machine learning model)|transformer]]s.
{{refend}}

== External links ==
<!--==== {{No more links}} ====
PLEASE BE CAUTIOUS IN ADDING MORE LINKS TO THIS ARTICLE. Wikipedia is not a collection of links nor should it be used for advertising.

Excessive or inappropriate links WILL BE DELETED.

See Wikipedia:External links & Wikipedia:Spam for details.

If there are already plentiful links, please propose additions or replacements on this article's discussion page, or submit your link to the correct category at the Curlie directory project (Curlie.org) and link back to that category using the {{Curlie}} template.
==== {{No more links}} ==== -->

* {{IEP|art-inte|Artificial Intelligence}}
* {{cite SEP |url-id=logic-ai |title=Logic and Artificial Intelligence |last=Thomason |first=Richmond}}
* [https://www.bbc.co.uk/programmes/p003k9fc Artificial Intelligence]. BBC Radio 4 discussion with John Agar, Alison Adam & Igor Aleksander (''In Our Time'', 8 December 2005).

{{Navboxes
|title = Articles related to Artificial intelligence
|list=
{{philosophy of mind}}
{{philosophy of science}}
{{Evolutionary computation}}
{{Differentiable computing}}
{{Computer science}}
{{emerging technologies|topics=yes|infocom=yes}}
{{Robotics}}
{{Existential risk from artificial intelligence}}
{{Glossaries of science and engineering}}
}}

<!--Please leave the empty space as standard.-->

[[Category:Artificial intelligence]]
[[Category:Cybernetics]]
[[Category:Computational neuroscience]]
[[Category:Computational fields of study]]
[[Category:Emerging technologies]]
[[Category:Formal sciences]]
[[Category:Unsolved problems in computer science]]

{{Sourceattribution|Artificial intelligence}}