Anthropic CEO outlines plan to slow AI development

Anthropic CEO Dario Amodei is arguing for a more deliberate approach to building advanced AI, aligning in broad terms with OpenAI CEO Sam Altman’s call to “pace the frontier.” The key unresolved question is what that slowdown would mean in practice, from model training schedules to safety testing and release controls.

The comments suggest top AI labs are moving from abstract safety concerns toward operational restraint as frontier models become more powerful. But any slowdown will be difficult to enforce without shared standards, because competitive pressure still rewards faster capability gains.