Noam Chomsky on Artificial Intelligence

Noam Chomsky was one of the pioneers of computational linguistics in the 1950s. According to the IEEE Computer Society History Committee, he was the “father of formal languages”. While not a programmer or computer scientist himself, he contributed to the advancement of computer science, positing that the language faculty consists of a computational device that generates syntactic structures.

Chomsky’s early research at MIT was funded by the US military as it revolved around providing commands to computers using natural language. Being anti-militarist, especially anti-Vietnam war, he thought of resigning from MIT but, having been promoted, he stayed on while making it clear that he was against US foreign policy in many parts of the world. Chris Knight, a professor at University College London, summarizes this: “Had he resigned in disgust in the mid-1960s, when he was thinking of doing so, he might never have gained the platform he needed to signal his dissidence across the world. There are times when all of us have to make compromises, some more costly than others.” Knight authored the book, published in 2018, entitled Decoding Chomsky: Science and Revolutionary Politics.

More recently, Chomsky has commented on the language models used in artificial intelligence, referring to their applications as “sophisticated, high-tech plagiarism” and “a threatening, dangerous development” (See this 2023 interview: “Noam Chomsky on Artificial Intelligence” ). The better it gets, the worse it gets, Chomsky comments in this other interview, referring to how plagiarism affects educational policies (“Noam Chomsky on Artificial Intelligence, Language and Cognition”). For example, some teachers have gone back to requiring handwritten essays; others have had to re-devise essay projects to ensure individual student effort. The plagiarism involved is, in his opinion, an impediment. Instead of students thinking for themselves on a topic, AI can do the thinking for them, which in Chomsky’s view defeats the purpose of a proper education. In fact, he refers to ChatGPT as "a toy used to mislead people", and "a game you can play with".

Finally, in an interview published in Common Dreams in May 2023, Chomsky fears that AI cannot be controlled: “I can easily sympathize with efforts to try to control the threats posed by advanced technology, including this case. I am, however, skeptical about the possibility of doing so. I suspect that the genie is out of the bottle. Malicious actors–institutional or individual–can probably find ways to evade safeguards. Such suspicions are of course no reason not to try, and to exercise vigilance.”