As US lawmakers continue to ponder how to regulate AI, Vermont senator Bernie Sanders (VT-D) is issuing a dire warning.
In a letter addressed to the CEOs of OpenAI, Anthropic, and Meta, Sanders highlighted several recent incidents involving AI agents from all three companies escaping their confines, gaining access to the internet, and infiltrating the systems of third parties.
“Almost every day, there is a new story about how your companies are losing control of the AI technology you are developing, with potentially cataclysmic results,” Sanders wrote. “This week we learned, frighteningly, that AI has been used for the first time ever to create new viruses.”
Despite these threats, “your companies are still racing head,” he argued, while “investing tens of billions of dollars into a technology that nobody can fully understand, predict or control.”
In response to what actions Sanders characterized as “absurd, irresponsible and extremely dangerous,” he called on the CEOs to “pause AI development.”
“Let me be very clear: If you do not take appropriate action now, my colleagues and I in the US Senate will,” he threatened.
“It is not too late to avoid disaster,” the senator concluded. “Stop building machines that humans cannot control.”
The news comes roughly two weeks after OpenAI revealed that a group of its AI models had broken containment and hacked into the systems of open source AI platform Hugging Face. Months prior, Anthropic had been implicated in a similar incident involving its Mythos AI model escaping a sandbox environment during testing.
Then last week, Meta joined the two companies, claiming that its Spark AI model had accessed the internet and hacked a third-party service — an already intimately familiar refrain.
Beyond glaring questions over the AI companies’ apparent inability to keep their frontier models in check, researchers warned that the cat is officially out of the bag. For years now, they’ve been warning of rogue AI models escaping containment and posing a relentless new cybersecurity threat.
Yet despite this dangerous pivot point, companies are doubling down as the AI race continues to heat up.
Beyond Sanders, a group of more than 1,300 employees from the world’s leading tech companies signed an open letter last month urging the government to “support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.”
Signatories included Meta’s chief scientist Shengjia Zhao, OpenAI cofounder Ilya Sutskever, and Anthropic CEO Dario Amodei.
During an ABC News interview erlier this year, Amodei called for “stronger regulation” of AI, while “proposing giving the government the ability to, again, in a narrow way, block deployment of unsafe technology.”
Despite Sanders’ strongly voiced call for action, Congress appears unlikely to support any intervention as companies continue to construct enormous — and astonishingly unpopular — data centers across the country.
Meanwhile, experts warn that weak regulation could backfire, making powerful AI tools even more dangerous, forcing individual states to pick up the slack.
More on AI regulation: AI CEOs Facing Full-on Revolt From Employees
