AI Out of Control? Whistleblower's Chilling Warning: 'Humans Don't Control AI'

Jacob Coxon, a former researcher at leading AI safety company Anthropic, has issued a stark warning that artificial intelligence companies do not fully control their own models. The whistleblower's claim, reported by Al Jazeera, highlights growing concerns that even the developers of advanced AI systems lack comprehensive oversight of their behavior.

Coxon's warning comes amid increasing scrutiny of AI safety practices, as companies like Anthropic, OpenAI, and Google DeepMind race to deploy increasingly powerful models. While specific details of his claims were not immediately available, his insider perspective adds weight to longstanding fears that commercial pressures may outpace safety measures. Anthropic has not yet issued a public response to the allegations.

The disclosure could intensify calls for stricter regulation and independent audits of AI systems. As governments worldwide debate AI governance, Coxon's warning may bolster arguments for mandatory transparency and control mechanisms. The incident also raises questions about the adequacy of internal safety protocols at major AI labs and could prompt further whistleblower disclosures.