An artificial intelligence researcher publicly resigned from Anthropic this week over concerns that the technology’s “out of control” progression could lead to widespread disruption that society is unprepared for, a warning that caused a stir online just as a push for more AI regulation has stalled in Washington.
Jacob Coxon’s posts on X circulated widely after he claimed that artificial intelligence companies are not “acting responsibly” with the technology and are instead racing to improve software models with no regard for the existential threat AI poses to the global economy and human civilization. Coxon, who said he worked as a pretraining researcher for three years at both OpenAI and Anthropic, added that AI would soon have the potential to “hack anything, revolutionize any field overnight, and acquire real power and resources.”
“The people building AI earnestly believe that it could kill us all by the end of the decade,” Coxon wrote. “This is not a marketing stunt. If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible — but I hear the same people express fear privately. No other human activity poses this level of danger.”
Two other leaders in the AI industry echoed Coxon’s concerns following his resignation. Evan Hubinger, a team leader at Anthropic, wrote on X that he believed there was a greater than 10% chance that AI could “kill all humans.” Samuel Marks, another team leader at Anthropic, said “the more senior the employee” at an AI company, “the more concerned they are” about the technology’s power.
Trending
OpenAI and Anthropic did not immediately respond to a request for comment on Coxon’s posts.
Coxon’s warning comes as public scrutiny grows over the rise of AI and the massive investment in data centers needed to power the technology. Lawmakers on both sides of the aisle have sought to regulate AI by introducing hundreds of bills, including measures to ban AI chatbots for minors, protect candidates for federal office from deepfakes and establish guardrails for the industry.
The proposals, however, have largely failed to materialize into legislation, leaving some lawmakers worried that Congress is missing a pivotal window to regulate AI.
Sen. Bernie Sanders (I-Vermont) has attempted to introduce several headline-grabbing measures in recent months, including a bill with Rep. Greg Casar (D-Texas) to ban “superintelligent” AI entirely and temporarily pause AI development while federal regulation is being discussed.
Rep. Lori Trahan (D-Massachusetts), who introduced a measure in June with Rep. Jay Obernolte (R-California) to establish a federal framework for AI, decided last week to not pursue a party leadership position in order to prioritize AI regulation.
President Donald Trump has embraced the AI industry during his second term, meeting with executives to discuss partnerships with the federal government and dismissing criticism of data centers. Trump has gone as far to say that communities that reject data centers will “end up being backwards and poor.”
“If they want to be successful and rich, with far lower taxes and jobs all over the place,” Trump wrote on Truth Social Aug. 31, “let Data Reign.”
Author
