Artificial intelligence company Anthropic announced Tuesday the public launch of its most advanced AI model yet, but with built-in restrictions preventing users from accessing cybersecurity functions that caused global concern earlier this year.
The new system, called Claude Fable 5, represents the startup’s most capable model available for widespread use, with the company highlighting its strengths in software development and data analysis tasks.
Previously, Anthropic had restricted the technology to approximately 200 organizations, including the U.S. government through its Glasswing program, following April’s announcement that the Mythos AI had identified thousands of software security weaknesses.
The broader release could help the $965 billion valued company maintain its competitive edge against rival OpenAI as both companies prepare for potential public offerings in the rapidly evolving AI sector.
According to the company, extensive testing was conducted to prevent users from circumventing the safety restrictions to perform prohibited activities.
“Let’s say I’m a college student asking the model like help me find cyber vulnerabilities on X package or code. The model would refuse and Fable 5 will fall back to Opus 4.8 for a response,” explained Dianne Penn, Anthropic’s head of product management, research and labs.
While Fable 5 carries a higher price point, Penn noted that early customer reports indicate it uses fewer tokens to complete tasks, ultimately reducing the total cost per assignment.
The company also announced that users who previously had access to the unrestricted preview version of Claude Mythos will have the option to upgrade to the new Claude Mythos 5.
Anthropic indicated plans to gradually broaden access through what it described as a more “systematic trusted-access program.”
Both AI models will be priced at $10 per million input tokens and $50 per million output tokens, the company stated.
