US Senator Bernie Sanders has called on the chief executives of Meta, OpenAI and Anthropic to immediately pause the development of advanced artificial intelligence systems, warning that the technology has reached a potentially dangerous threshold.
In a letter addressed to Meta CEO Mark Zuckerberg, OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei, the Vermont senator argued that the rapid advancement of AI poses serious risks to humanity and urged the companies to slow development until stronger safety measures are established.
“Mr Altman, Mr Amodei and Mr Zuckerberg: In the interest of humanity, stand by your words. Pause AI development,” Sanders wrote, warning: “Stop building machines that humans cannot control.”
Sanders cited recent reports suggesting that advanced AI systems have been used in the development of potentially dangerous biological agents. He warned that if such capabilities fell into the wrong hands, they could potentially be used to create bioweapons capable of causing mass casualties.
The senator also pointed to disclosures from major AI companies that their models had been involved in hacking other organisations’ computer systems, describing the incidents as evidence that developers were increasingly struggling to maintain control over increasingly autonomous AI systems.
“AI capabilities HAVE reached a critical threshold,” Sanders declared, arguing that recent incidents demonstrate that the conditions under which companies had previously pledged to slow or halt development may already have been met.
Sanders warned that the US Senate would move to regulate the industry if companies continued deploying increasingly powerful AI systems without adequate safeguards.
His intervention comes amid growing concerns among scientists, technology experts and policymakers over the pace of frontier AI development.
More than 1,300 scientists, engineers and technology professionals have reportedly signed an open letter calling for greater international coordination to slow the development of advanced AI models and give governments time to establish appropriate safety regulations.
The signatories argued that intense competition between technology companies and countries is encouraging developers to move faster than regulators can identify and address emerging risks.
The senator also referenced previous commitments made by AI companies to pause or slow development if their systems reached certain risk thresholds.
Anthropic has said it would pause the scaling or delay deployment of new models if its ability to expand AI capabilities outpaced its ability to comply with established safety procedures.
OpenAI has similarly committed to halting further development if its AI capabilities reach a defined “critical” threshold until appropriate safeguards are implemented.
Sanders argues that recent autonomous hacking incidents and other developments indicate that the threshold has now been reached.
The senator has previously advocated a temporary moratorium on the construction of new data centres in the United States, arguing that a pause would give policymakers time to establish stronger federal safeguards for artificial intelligence.
In March, Sanders and Congresswoman Alexandria Ocasio-Cortez introduced legislation seeking to temporarily halt new data-centre construction while the government develops a comprehensive regulatory framework for the rapidly expanding AI industry.
Read the full statement here:
Bernie Sanders Demands AI Development Pause, Warns of ‘Machines Humans Cannot Control’
