Anthropic releases Opus 5 promising Fable 5-like capabilities

Summary

Anthropic launched Claude Opus 5; this model rivals the flagship Fable 5's capabilities while offering stronger cybersecurity safeguards and enhanced coding performance.

Why it matters

This release acts as a strategic response to recent government regulations and OpenAI's security incidents, positioning Anthropic as a competitive corporate AI provider. By emphasizing compliance and reduced misuse risk, Opus 5 maintains cost parity with its predecessor while aiming to outpace rivals like GPT-5.6, thereby seeking to regain trust.

Highlights

• Opus 5 claims to be 'close' to Fable 5's abilities, particularly in complex coding tasks.
• Anthropic asserts that it provides stronger cybersecurity measures and high compliance to prevent misuse.
• Pricing remains the same as Opus 4.8; this makes it half the cost of Fable 5 and slightly cheaper than OpenAI's GPT-5.6.

Critical take

The article does not provide concrete performance benchmarks for Opus 5, making the 'close to Fable 5' claim difficult to verify. Moreover, reliance on government testing could delay broader adoption, and the optional 'fast mode' pricing may create a tiered experience for users.