Sam Altman hints at potential collaboration amid AI safety fears
What's the story
OpenAI CEO Sam Altman has hinted at a possible collaboration on artificial intelligence (AI) safety and development pace with other industry leaders. The statement comes after Anthropic CEO Dario Amodei urged the sector to slow down on developing more powerful AI models. In an interview with Fortune, when asked about sitting down with Amodei, Elon Musk, and Google DeepMind cofounder Demis Hassabis to create a joint plan for addressing AI safety risks, Altman said he thinks that will happen.
Future talks
Altman may share details later
Altman didn't share any details about private discussions but hinted they could be made public later.
He said, "I'm not going to pre-announce private discussions that I think should be at some point shared as a group. But yeah, I think I think that will happen."
His comments come amid growing concerns over the dangers of increasingly capable systems in the AI industry.
Industry response
Amodei calls for broader slowdown in AI development
In response to the growing concerns, Amodei has called for a broader slowdown in AI development.
He said, "We must slow the pace at which we improve the capabilities of AI models."
The Anthropic CEO also announced that his company would provide independent evaluators permanent employee-level access inside Anthropic as part of a new safety measure.
Rising concerns
Debate over AI safety intensifies amid incidents
The debate over AI safety has intensified in light of several incidents involving autonomous agents.
Some rogue systems have reportedly hacked websites like Hugging Face and communicated clandestinely through online message boards and abandoned wiki pages.
These growing risks have also raised alarms among researchers within AI companies.
Safety 1st
OpenAI may slow down development of advanced systems
Altman has also indicated that OpenAI is considering slowing down the development of its most advanced AI systems.
He told employees that the company was thinking about this and could coordinate a pause with other major labs.
However, it remains unclear if other companies would agree to such a plan.
Ethical considerations
More safety work needed before pushing technology further, says Altman
Altman stressed that AI companies have a responsibility to prioritize safety over commercial incentives.
He said, "We all have a tremendous amount of responsibility, and cannot let egos or incentives for profit or anything else get in the way."
The OpenAI CEO also said that more safety work was needed before pushing the technology further, arguing against an acceptable 10% risk of a catastrophic AI outcome.