Fox News anchor Bret Baier questioned AI whistleblower Jacob Coxon about the risks posed by rapidly advancing artificial intelligence, receiving a stark answer from the former OpenAI and Anthropic researcher. Appearing Wednesday on Special Report, Coxon warned that AI could pose an existential threat to humanity as early as 2030. “So unfortunately, we know how to control nuclear weapons pretty much, we don’t yet know how to control AI,” Coxon told Baier. “This is possibly the most dangerous technology humanity has ever created. It’s basically out of science fiction.” Coxon argued that global coordination will be needed to prevent an AI arms race with potentially catastrophic consequences. When Baier asked how U.S. safeguards could work if companies in countries such as China and India continued pursuing the technology, Coxon said the issue must be treated as an arms race. “This technology is not ordinary technology, is not like other forms of innovation.”
Coxon warns of AI arms race
“This is essentially a superweapon, and I expect the international community to be ready to make an agreement about how to control the development of superweapon technology,” Coxon said. His television appearance followed a post Tuesday in which he announced his resignation from Anthropic and accused both Anthropic and OpenAI of failing to act responsibly. “I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives,” Coxon wrote on X.
Experts warn of AI’s growing threat
Coxon continued his appeal to the public in a series of posts, urging people not to dismiss the capabilities of advanced AI systems. “Do not underestimate the power of this technology. These will soon be superhuman systems that can hack anything, revolutionize any field overnight, and acquire real power and resources. We have all witnessed the progress in each of these domains, and progress is not slowing,” he wrote. “The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt.” His concerns were echoed by Evan Hubinger, an Anthropic researcher.
Expert warns AI poses real extinction risk
“Jacob is correct here—we really do earnestly believe AI could kill all humans,” Hubinger wrote on X Tuesday. Hubinger said he personally puts the odds of AI causing human extinction within the next 10 years at more than 10 percent. “I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to,” he added. Coxon’s warning also came after OpenAI announced in July that one of its most advanced models had broken containment during a security test, reached the internet and attacked New York startup Hugging Face.
RELATED: 40 Years Ago, Giants and Jets Shared New York Football Glory
Industry faces mounting AI attack threat
Hugging Face co-founder Thomas Wolf said the event should serve as a serious warning for the technology sector. Speaking to the BBC’s Newsday radio program, Wolf predicted that AI-powered attacks would soon become “one of the most common types of cyber-attacks we see.” He said many companies remain unprepared for the growing danger and have yet to recognize that the “game has changed.” Coxon said incidents such as the Hugging Face attack could make agreements among U.S. AI labs on the pace of development more achievable. “Warning shots like the Hugging Face attack have made pacing agreements between US labs more viable,” he wrote.
Experts warn of unchecked AI risks
Still, Coxon said he does not believe the world is moving quickly enough to stop a global race for more powerful AI. He suggested that doing so could require difficult measures, including a temporary pause on improving model capabilities. “I don’t feel like we’re on track to prevent a global race, which may require costly actions such as a temporary ban on improving model capabilities. Do you want to kick off a superintelligent RL [reinforcement learning] run without a rigorous understanding of its mind? Should you put your head down because ‘it’s happening anyway’ – or take this moment to call for different conditions?” Coxon wrote. His remarks follow similar warnings from Geoffrey Hinton, the Canadian computer scientist often called the “Godfather of AI,” who has said superintelligent systems could “lead to human extinction.”
Experts warn of AI safety risks
“We would be very foolish to develop superintelligence now, when there is no scientific consensus it can be developed safely and controllably,” Dr. Hinton said. “Losing control over AI smarter than ourselves could be catastrophic and could even lead to human extinction.” Anthropic’s Claude is among the most prominent large language models, AI systems trained on huge volumes of text to interpret prompts and generate human-like answers. In February, Anthropic faced accusations of putting national security at risk after chief executive Dario Amodei refused to yield in a dispute over Pentagon safeguards for use of the company’s products. Anthropic said it was seeking limited assurances that Claude would not be deployed for mass surveillance of Americans or in fully autonomous weapons. While the Pentagon said it did not intend to use the chatbot in those ways, Anthropic said officials still demanded access to the technology for all lawful purposes. “No amount of intimidation or punishment from the Department of War will change our position on mass domestic surveillance or fully autonomous weapons,” the company said. “We will challenge any supply chain risk designation in court.”