RECKLESS: Anthropic Whistleblower Warns AI Firms 'Don't Fully Control' Their Models...

A former Anthropic safety researcher testified before lawmakers in New York that AI companies lack genuine control over their models' behavior, describing the industry's current safety practices as reckless. Jacob Coxon's warning highlights a growing divide between public assurances from AI firms and behind-the-scenes concerns among employees about whether these systems can be reliably contained or predicted once deployed at scale.