MENLO PARK, California- Meta Chief Executive Mark Zuckerberg has rejected calls for technology companies to coordinate a slowdown in artificial intelligence development, arguing that individual firms already have strong incentives to build safe systems.
Zuckerberg said AI developers should take responsibility for assessing their models and delaying products that do not meet safety standards instead of relying on new government regulations or an industry-wide pause.
He argued that companies risk losing users and facing legal liability if they release AI systems that are unsafe or fail to operate as intended.
“I think every lab has a responsibility to make sure that what they’re building is safe,” Zuckerberg said.
The Meta chief cited the company’s decision to delay the release of its Muse AI agent for several months while engineers worked to improve its security.
He said Meta did not ask competing companies to suspend their own projects during that period, maintaining that each developer should make independent decisions about its products.
Zuckerberg also supported the use of independent evaluators to assess advanced AI systems. He said involving a diverse group of external reviewers could help companies identify risks before releasing products to the public.
His remarks came amid a widening disagreement among technology executives over the speed at which increasingly powerful AI systems should be developed.
Anthropic Chief Executive Dario Amodei has called for stronger coordination and greater scrutiny of advanced models, warning that rapidly developing AI could pose serious risks.
OpenAI Chief Executive Sam Altman and other technology figures have also backed measures aimed at improving security, external testing and cooperation between leading AI developers.
Nvidia Chief Executive Jensen Huang has opposed slowing the technology’s development, saying innovation and safety can proceed together. He has argued that companies should stop or delay individual products when safety concerns emerge instead of imposing broader restrictions across the industry.
US President Donald Trump has also opposed efforts to slow American AI development, arguing that excessive restrictions could allow China to gain an advantage in the global technology race.
Zuckerberg’s position places Meta among companies supporting continued rapid development combined with internal safety controls and external evaluations.
Meta has invested heavily in AI infrastructure and the development of models used across Facebook, Instagram, WhatsApp and its other products.
