September 10, 2026

AI Regulation Urged as Expert Warns of Potential Threats

Jacob Coxon, a former Anthropic researcher, has raised concerns about the potential dangers of artificial intelligence. Speaking on ‘Special Report,’ he emphasized the importance of regulation to avoid catastrophic consequences.

Coxon announced his resignation on X, citing the irresponsible actions of companies in the pursuit of self-improving AI as a major concern. He warned that this technology poses a significant risk to humanity within the next decade.

During his interview with Bret Baier, Coxon highlighted the solvability of the problem. He stressed the need for careful and deliberate regulation, noting that the rapid advancement in AI is leaving insufficient time to address these risks appropriately. The global demand for regulation is seen as crucial to prevent a race to potential extinction.

Evan Hubinger, the alignment science lead at Anthropic, echoed Coxon’s sentiments on X. He acknowledged the possible risk of AI causing human extinction and expressed concerns about the lack of a clear plan to ensure AI alignment in the future.

Hubinger mentioned that while current AI models present low risk, the advent of self-improving AI could significantly increase the threat. The industry anticipates that AI could become dangerously self-improving within a short timeframe, potentially leading to a major takeover before the decade ends.

Coxon also addressed the international arms race aspect of AI, labeling the technology as a ‘superweapon.’ He called for global cooperation to regulate AI technologies, emphasizing that such collaboration is vital to avoid disastrous outcomes.

He referred to calls from technology leaders for a slow-down agreement, noting that these are serious proposals and not merely marketing tactics. Despite the promise of significant discoveries, Coxon urged moderation in the rapid development of AI.

When asked about the possibility of resolving these issues, Coxon expressed confidence. He believes that through concerted effort and avoiding a pessimistic outlook, the risks posed by AI can be mitigated effectively.

Fox News Digital attempted to reach Anthropic for comment. Sydney, an editor with Fox News Digital, is available for story tips via email at [email protected].

TAGS: