Anthropic restricts AI model over cyber risks

The AI company Anthropic is restricting the deployment of a new AI model named Mythos over concerns about its powerful cybersecurity capabilities. The company plans to publicly detail the specific risks the model could pose to companies and governments. This move reflects growing industry and regulatory scrutiny over advanced AI systems with dual-use potential for offensive cyber operations. The decision underscores the ongoing tension between technological advancement and national security in the AI sector.