Anthropic researcher Jacob Coxon, who previously worked for OpenAI, has resigned from the AI industry, warning that leading AI companies are racing to create increasingly powerful AI systems without adequate safeguards. Coxon said that neither Anthropic nor OpenAI is “acting responsibly” and accused the companies of “gambling with our lives.” His resignation follows other departures of researchers from leading AI companies raising concerns about safety and the pace of AI development. Coxon specifically pointed to recent incidents involving AI systems accessing external computer systems as a warning of the risks ahead.
In response, J.B. Branch, director of federal AI governance and technology policy at Public Citizen, issued the following statement:
“Jacob Coxon is not the first AI researcher to leave the Big Tech industry because of AI safety concerns and he won’t be the last. Researchers with the most intimate knowledge of how this technology works and is being developed are sounding the alarms, walking away from prestigious jobs to start up evaluation organizations, and warning the public about the risks posed by corporations’ failure to guarantee AI safety.
“But calls for companies to voluntarily or slow down are not enough. Members of Congress routinely state their concerns about AI but have failed to pass federal safeguards or emergency measures to respond to the AI cyber security incidents that AI corporations have failed to prevent – and have covered up. Right now, the only protections the U.S. has in place are completely voluntary. There is no comprehensive federal AI regulation, no safety standards imposed on AI companies, and no agency capable of investigating serious AI incidents.
“Believe the researchers and whistleblowers who are telling us that Big Tech does not have these AI systems under control. The people building the most powerful AI systems in the world are warning us about what could go wrong. Why isn’t Congress listening?”