Anthropic Launches Claude Opus 5.5 Amid Calls for AI Caution
Despite recent calls from its CEO for greater caution in the development of artificial intelligence, Anthropic continues to advance its AI models. The American company launched Claude Opus 5.5 on Tuesday, September 22, marking the debut of its new Claude 5.5 generation.
A Strategic Timing
The timing of this release is significant. Just days prior, Anthropic CEO Dario Amodei had urged industry players to "slow down the frontier" to better align the rapid progression of AI capabilities with the development of robust safety measures. Claude Opus 5.5 is the first model released by Anthropic since this public stance.
Enhanced Performance and Reduced Costs
Anthropic states that Claude Opus 5.5 achieves performance levels comparable to Claude Fable 5.1 on most tasks, while requiring 40% lower operational costs than its predecessor, Opus 5.
- Pricing: The new model is priced at $4 per million input tokens and $20 per million output tokens, down from the previous rates of $5 and $25, respectively.
- Key Use Cases: The model is specifically optimized for agentic coding, complex professional tasks, and computer use.
- Speed and Efficiency: Anthropic highlights a generation speed that is over 30% faster than Opus 5, along with reduced token consumption for certain tasks.
Strengthened Safety Measures
In parallel with performance improvements, Anthropic has reinforced the safety protocols surrounding the model. Claude Opus 5.5 has been evaluated by external organizations, including Frontier Design and METR. It also benefits from protective measures previously reserved for the company’s most advanced models.
Future Roadmap
Anthropic plans to complete this new model family with the release of Claude Sonnet 5.5 and Claude Haiku 5.5 in the coming weeks. The launch of Opus 5.5 illustrates the ongoing dilemma faced by major AI laboratories: the challenge of strengthening safety mechanisms while simultaneously improving the efficiency and capabilities of their models.
