Should we pause development of AI systems smarter than GPT-4 for six months?
Mar. 30, 2023.
2 min. read Interactions
Should we pause development of AI systems because of risks to society and humanity? Experts are not in agreement.
A “call for a pause” open letter sent out Wednesday warns of an “out-of-control race to develop and deploy ever-more-powerful AI systems.”
The letter, by Future of Life Institute, an organization focused on technological risks to humanity, warns that “AI systems with human-competitive intelligence can pose profound risks to society and humanity, as shown by extensive research.”
The letter calls on “all AI labs to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4.” As of early Thursday, the letter has been signed by verified AI experts at 1377 AI labs.
One AI expert not on that list: Dr. Ben Goertzel, CEO of SingularityNet and an AI scientist noted for his work in AGI (artificial general intelligence). “I don’t see how pausing training of bigger LLMs [large language models] for six months is going to suddenly cause the corporate honchos and tech bros to start pushing toward UBI for the poorest nations in the world, in anticipation of the difficulty pre-Singularity AI will cause to their economies,” he told Mindplex.
“I think the justification for pausing tech dev due to worries should need to be quite strong. Just like I think the justification for suspending freedom of speech should be quite strong. OK, you ban shouting fire in a crowded theater. But you don’t ban speech that seems like it might indirectly cause someone to do something illegal.
“Similarly, you take action against an LLM that demonstrably turns everyone who talks to it into murderous psychopaths, or a 3D printer whose express purpose is to cheaply 3D print bombs out of ordinary kitchen ingredients. But not against exciting new technologies with complex mixes of good and bad aspects and no immediate huge deadly threat associated with them.”
Other leading AI experts who have refused to sign include OpenAI CEO Sam Altman and Yann LeCun, Chief AI Scientist at Meta, who explained: “The year is 1440 and the Catholic Church has called for a 6 months moratorium on the use of the printing press and the movable type,” he said. “Imagine what could happen if commoners get access to books! They could read the Bible for themselves and society would be destroyed.”