Meta CEO Mark Zuckerberg said he opposes the pursuit of a single artificial general intelligence, describing the idea as a "pretty big turnoff" in a new interview with YouTuber Kane Sutter. Zuckerberg criticized unspecified tech industry rivals who talk about building "this 'one true AI,'" saying, "It's almost as if they kind of think they're creating God or something."
He argued that a single all-purpose AI would not be practical because people have different needs and interests. Instead, he said, many different AIs performing different tasks would be required. Zuckerberg also criticized closed AI platforms and promoted open-source AI, saying it would let people create AIs tailored to their own needs or those of others.
Meta's Position in the AI Race
The comments come as Zuckerberg tries to position Meta as a leader in AI development. The company has faced recent setbacks, including Apple declining to integrate Meta's AI systems and Facebook being filled with AI-generated content.
Competing Views on AGI's Timeline
Zuckerberg's remarks contrast with other industry figures. OpenAI CEO Sam Altman has said AGI is coming in the "reasonably close-ish future" and is prepared to spend heavily to achieve it. OpenAI leadership had previously been confident enough about AGI's arrival that the company seriously considered auctioning its future AGI to governments including Russia and China. Shane Legg, chief AGI scientist at Google DeepMind, has put the probability of AGI arriving by 2028 at 50 percent.
Others are more skeptical. Grady Booch, an IBM Fellow and chief scientist for software engineering, told AI critic Gary Marcus last year that AGI will never happen. Booch described himself as a contrarian on such predictions, citing his background as a historian of computing and what he called the field's "hyperbolic optimism."
Whatever the predictions, the future of AGI remains undetermined.
Editorial Notes