Skip to content
The Chat Archive
ChatGPT (model unspecified)· OpenAI ShareLM (browser-extension contributions)

After the many efforts to create AI (in the 70's 90's etc.) there was a feeling that the…

  1. User

    After the many efforts to create AI (in the 70's 90's etc.) there was a feeling that the goalpost keeps moving and "real AI" is always what current AI can't do, as AI solves more and more complex problems but none of them suffice. Which citations reflect that?

  2. Assistant

    Great question — you’re describing what’s often called the “AI effect”: once a capability is achieved, people stop calling it "AI" and redefine "real AI" as whatever remains unsolved. This shifting-goalpost idea has been discussed for decades by AI researchers, philosophers, and critics. Here are some key citations/quotes that reflect it:

    Classic Sources

    Pamela McCorduck (1979, Machines Who Think)

    "Whenever a machine is made to do something — play checkers, prove mathematical theorems, compose music — there’s a chorus of voices saying, 'that’s not thinking.'"

    Larry Tesler (often paraphrased, 1980s)

    “AI is whatever hasn’t been done yet.” (Sometimes attributed as Tesler’s Theorem: ‘AI is whatever doesn’t work yet.’)

    Douglas Hofstadter (1985, Metamagical Themas)

    "As soon as it works, no one calls it AI anymore."

    John McCarthy (coiner of the term "AI") Wrote in the 1980s about the “AI effect”:

    “As soon as it works, no one calls it AI.”

    Academic/Analytical Sources

    Daniel Crevier (1993, AI: The Tumultuous History of the Search for Artificial Intelligence)

    “Once some piece of software or hardware is able to do something — say, chess playing — people stop saying it is AI.”

    Nils J. Nilsson (1998, Artificial Intelligence: A New Synthesis) Explicitly discusses the “AI effect” and how progress is dismissed once incorporated into standard tools.

    Nick Bostrom (2014, Superintelligence)

    "Many activities once considered to require 'intelligence' are now routine for machines... each success narrows the field of what counts as AI."

    Summary Term

    The “AI Effect” (sometimes also called the moving target problem):

    "AI is defined as whatever machines can’t yet do." (Common shorthand in both technical and popular discussions.)

    Would you like me to compile a clean reference list (APA/MLA style) of these sources so you can directly cite them in academic work?