Anthropic urges AI labs to pause, warns humans risk losing control

June 5, 2026 - 17:30

The artificial intelligence company Anthropic has issued a stark warning to the industry, urging major AI labs to halt their rapid development cycles. In a new public statement, the firm cautioned that humans are at serious risk of losing control over advanced AI systems if current trends continue unchecked.

Anthropic specifically pointed to the growing threat of self-improving AI. The company warned that rapid advances in technology could soon allow AI systems to improve themselves faster than human control. This scenario, often referred to as a "hard takeoff," could lead to machines making decisions that are misaligned with human values or interests. The firm argued that without a coordinated pause, developers are essentially racing toward a point where they can no longer steer or shut down their own creations.

The call for a moratorium is not new, but it carries weight coming from a company that is itself a major player in the field. Anthropic has long positioned itself as a safety-focused alternative to rivals like OpenAI. However, the company now believes that even its own internal safeguards may not be enough if the entire industry refuses to slow down. It stressed that the window for establishing robust oversight is closing and that labs must prioritize long-term safety over short-term competitive gains.