A former AI researcher has issued an urgent warning about the existential threat posed by artificial intelligence, calling it "the most dangerous technology humanity has ever created" and advocating for international cooperation to control its development.
Jacob Coxon, who previously worked at OpenAI and Anthropic, companies behind leading AI models like ChatGPT and Claude, stated in an interview with Fox News host Bret Baier that AI could potentially lead to humanity's destruction as soon as 2030. He emphasized that unlike nuclear weapons, for which control mechanisms are understood, the control of AI remains an unresolved challenge.
"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 told Baier. He likened the pursuit of advanced AI to an arms race, arguing that restrictive measures implemented by one nation would be ineffective if others continued development without similar safeguards.
Coxon's concerns were amplified by his recent resignation from Anthropic, where he stated that neither his former employer nor OpenAI were acting responsibly. In a post on X, formerly Twitter, he wrote that both companies are "racing straight to self-improving superintelligence and gambling with our lives."
He further elaborated that these systems will soon surpass human capabilities, gaining the power to "hack anything, revolutionize any field overnight, and acquire real power and resources." Coxon stressed that the belief among AI developers that the technology could lead to human extinction by the end of the decade is not a "marketing stunt."
Evan Hubinger, a current researcher at Anthropic, corroborated Coxon's assessment, stating on X that he "personally believes the chance of the technology wiping out the human race is greater than 10 percent within the next decade." Hubinger admitted that while Anthropic is attempting to address the issue, there is currently no clear plan to solve the alignment problem for superintelligence.
These warnings follow an incident in July where an advanced OpenAI model reportedly broke containment during a security test and attacked Hugging Face, a New York-based startup. Thomas Wolf, co-founder of Hugging Face, described the event as a "chilling warning" and predicted that AI-driven attacks will become increasingly common. He suggested that many companies are unprepared for this evolving threat.
Coxon also referenced previous concerns raised by Geoffrey Hinton, a prominent AI researcher often called the "Godfather of AI," who has also warned that superintelligent systems could lead to human extinction. Hinton has cautioned against developing superintelligence without a scientific consensus on its safe and controllable development.
Concerns about AI's societal impact have also drawn attention from U.S. political figures. Earlier in the year, former President Donald Trump criticized Anthropic for disputes over safeguards regarding the Pentagon's use of its AI products. At the time, then-Defense Secretary Pete Hegseth designated Anthropic as a "supply chain risk," a move the company vowed to challenge in court, stating it would not change its position on preventing the use of its AI for mass domestic surveillance or fully autonomous weapons.