Anthropic co-founder Jack Clark has called for the establishment of a safety framework for artificial intelligence, warning that rapid advancements in the field could pose an existential threat to humanity. Clark expressed these concerns in a recent interview with Fox News, asserting that proactive regulation is necessary to prevent potential accidents and ensure public confidence in the technology.
Clark emphasized that a "sensible, common-sense safety framework" is needed for the AI industry. He argued that waiting to impose restrictions until after a major incident would be detrimental, potentially harming the United States' competitive edge against China in AI development. "The worst time to react to an exponential [threat] is when it's too late," he stated.
The executive rejected suggestions that Anthropic's stance is motivated by financial interests, calling such theories "absurd." He noted that Anthropic has previously paused development on its AI model, Claude, to focus on safety measures. However, he stressed that companies cannot address these risks alone and require industry-wide collaboration and a broader policy conversation.
Clark also urged former President Donald Trump to raise AI safety concerns during discussions with Chinese President Xi Jinping, framing the issue as a critical moment for human agency in shaping the future.
President Trump, however, has dismissed concerns about AI as "sick conspiracies." On Truth Social, he stated that the U.S. already has the necessary "guardrails" with a "strong and smart (High IQ!) president." He specifically named Anthropic CEO Dario Amodei as an "AI alarmist" and accused him of posing as a "perfect little angel." Trump asserted that his administration has prevented AI companies from engaging in "bad, or potentially bad, ‘things.’"
Trump has consistently argued that halting U.S. AI development would primarily benefit China. He suggested that anti-data center protests are being instigated by Chinese officials, aiming to slow U.S. progress. "Whoever wins AI, wins," Trump wrote, emphasizing the U.S. lead over China and warning "Conspiracy Theorists, Treasonists, Traitors, and Leakers, beware!"
He further characterized the widespread concern around AI as a "hoax," questioning why industry leaders would advocate for regulations that could lead to their "oblivion and bankruptcy." Trump described AI and data centers as a "Greatest Economic Development Engine in History," surpassing oil, gold, diamonds, and the internet, and stated it would not be stopped during his term.
Others within the Trump administration and associated circles have echoed skepticism. Vice President JD Vance expressed bewilderment at AI companies seeking government regulation, likening it to a "Trojan horse." Former Trump AI and crypto czar David Sacks voiced concerns about "regulatory capture," where large corporations could use new regulations to disadvantage smaller competitors who may not be able to afford compliance.
Clark countered these arguments by noting Anthropic's past investments in safety and stressing the need for a unified approach. "We can't act alone," he said. "We need to act with industry [executives] to develop a common set of standards, and we need ultimately there to be a larger policy conversation about what do we think are the common rules of the road here for everyone building this technology."