Anthropic’s Claude aims to be a safe, high-powered AI coworker for professionals

In an exclusive Tom’s Guide interview, Claude AI’s product lead explains Anthropic’s safety-first approach to building a powerful, interpretable AI that can act as a professional coworker. The discussion covers Claude’s emphasis on safety over hype, its browser-related capabilities, and how the company balances growth with guardrails to reduce risks like prompt injections, all while pursuing practical, work-focused use cases rather than flashy features.