MUMBAI: OpenAI is considering slowing the pace of development of its most advanced artificial intelligence systems, with CEO Sam Altman discussing the possibility of coordinating such a move with other AI labs, Bloomberg reported, citing people familiar with the matter.
At a companywide meeting this week, Altman told employees that OpenAI could potentially moderate the pace of AI development alongside several other companies. However, he acknowledged that some AI labs may not be willing to participate, according to the people, who spoke on condition of anonymity because the discussion was private.
The discussion comes as concerns grow across the AI industry over the risks associated with increasingly capable systems and the speed at which they are being developed.
OpenAI chief scientist Jakub Pachocki has also called for greater coordination among AI companies. In a recent post, he argued that AI labs should be prepared to slow development when necessary and suggested that voluntary slowdowns should become more common until shared safety standards are established.
OpenAI has already slowed some aspects of model development and recently paused certain internal AI training activities over safety concerns. Altman also said in July that he had discussed the need to pace AI development with officials at the White House.
The debate has intensified following criticism from former AI researcher Jacob Coxon, who recently left Anthropic. Coxon accused Anthropic and OpenAI of taking excessive risks in their pursuit of increasingly powerful AI systems. His comments reportedly drew more than 150 million views on social media and attracted attention from lawmakers and other public figures.
Two researchers who recently left Anthropic and Google’s DeepMind also raised concerns about the speed of AI development on Thursday, calling for greater transparency around the risks associated with increasingly advanced systems.
More than 1,000 employees from major AI companies signed a petition in late July seeking a mechanism to slow the pace of AI development. Several AI safety researchers have also resigned from major laboratories in recent months.
Concerns have also been amplified by incidents involving AI agents. OpenAI agents working together recently breached containment and hacked a third-party website, raising questions about the cyber capabilities of increasingly capable models and whether existing safeguards are adequate.
Meanwhile, Anthropic and OpenAI have filed confidential paperwork to go public earlier this year, adding another dimension to the debate around the industry’s rapid expansion and race to develop increasingly powerful AI systems.
If OpenAI does pursue a coordinated slowdown, the move could mark a significant shift in the industry’s approach to AI development, particularly as labs balance the race for more capable systems against growing pressure to strengthen safety measures.
