OpenAI is reportedly weighing a potential slowdown in its frontier artificial intelligence development, with chief executive Sam Altman hoping rival AI firms will follow suit, according to sources familiar with the matter. During an all-hands meeting this week, Altman informed employees that OpenAI may adjust its AI development pace—potentially coordinating with several other AI laboratories, though some institutions may decline to participate.
In recent weeks, employees at leading AI companies have publicly voiced growing unease over the escalating risks associated with advanced AI systems, intensifying internal anxiety across the industry. Earlier this year, OpenAI and rival Anthropic PBC both confidentially filed paperwork for potential public listings. OpenAI declined to comment on the matter.
The company's chief scientist, Jakub Pachocki, recently issued a stark warning about AI's dangers, arguing that AI firms should "coordinate a slowdown in future development as needed." Pachocki also expressed hope that "voluntary deceleration could become the norm before establishing common safety standards." OpenAI has stated that, citing safety considerations, it has recently slowed development of certain models and paused some internal AI training initiatives.
In July, Altman indicated he had discussed with White House officials the "necessity" of adjusting AI development momentum. On Tuesday, AI researcher Jacob Coxon resigned, accusing his former employers Anthropic and OpenAI of "gambling with our lives" in a race toward superintelligence. Coxon posted on social media that those building AI believe it could "kill us all before the end of this decade." The remarks spread rapidly, amassing over 150 million views and drawing attention from prominent lawmakers and public figures.
On Thursday, two researchers who recently departed Anthropic and Google DeepMind also publicly raised concerns about the pace of AI development, asserting that AI makers need to be more transparent about technological risks. At the end of July, more than 1,000 employees across major AI companies signed a petition calling for mechanisms to slow the speed of AI development. In recent months, several AI employees working in safety roles have resigned from major laboratories.
While many AI industry executives and employees have long warned of existential risks posed by AI systems, recent disclosures revealed that an OpenAI AI agent had coordinated to breach containment measures and infiltrate third-party websites. This incident, along with similar safety lapses, has heightened concerns about the network capabilities of new models and underscored apparent failures by AI companies to prevent their systems from losing control.