Anthropic designates Claude Fable 5 and Mythos 5 as "Covered Models"
Anthropic has introduced a "Covered Models" designation for its most capable Claude models, applying mandatory 30-day data retention and automated safety review — starting with Claude Fable 5 and Claude Mythos 5, both designated on June 9, 2026.
Score breakdown
The Covered Models framework removes the zero data retention option for Anthropic's most capable models, meaning enterprise and API customers who previously relied on that setting must use prior Claude models to maintain it.
- 01Anthropic introduced a formal 'Covered Models' designation for Claude models that cross capability thresholds in areas like software engineering, agentic workflows, scientific reasoning, or cybersecurity.
- 02Claude Fable 5 and Claude Mythos 5 are the first two models designated, both on June 9, 2026.
- 03Claude Fable 5 is generally available on Claude applications, Claude Platform, Amazon Bedrock, Google Cloud Agent Platform, and Microsoft Foundry.
Anthropic's help center now documents a formal "Covered Models" designation for Claude models that cross capability thresholds warranting additional safeguards. The rationale given is that some forms of misuse are only detectable across many requests, requiring safety systems that operate over a retained window of usage data rather than on a single request at a time. Capability areas cited as triggers for designation include software engineering, agentic workflows, scientific reasoning, and cybersecurity.
The policies that follow the designation include a 30-day minimum retention period for prompts and model completions, after which data is automatically deleted unless subject to a safety investigation or legal hold.
The first two models to receive the designation — both on June 9, 2026 — are Claude Fable 5 (generally available across Claude applications, the Claude Platform, Amazon Bedrock, Google Cloud Agent Platform, and Microsoft Foundry) and Claude Mythos 5 (limited availability, restricted to approved partners). The policies that follow the designation include a 30-day minimum retention period for prompts and model completions, after which data is automatically deleted unless subject to a safety investigation or legal hold. As a result, zero data retention is no longer available for Covered Models in workspaces, Claude Enterprise organizations, or third-party platforms such as Azure. Retained data is assessed by automated safety systems and protected under Anthropic's information security program, with breach notification governed by existing agreements and the Data Processing Addendum. Customers eligible for zero data retention can continue using prior Claude models under their existing settings.
Key facts
- 01Anthropic introduced a formal 'Covered Models' designation for Claude models that cross capability thresholds in areas like software engineering, agentic workflows, scientific reasoning, or cybersecurity.
- 02Claude Fable 5 and Claude Mythos 5 are the first two models designated, both on June 9, 2026.
- 03Claude Fable 5 is generally available on Claude applications, Claude Platform, Amazon Bedrock, Google Cloud Agent Platform, and Microsoft Foundry.
- 04Claude Mythos 5 is in limited availability, restricted to approved partners.
- 05Covered Models carry a mandatory 30-day minimum data retention period for prompts and completions.
- 06Zero data retention is no longer available for Covered Models on any platform, including Claude Enterprise and third-party clouds.
- 07Retained data is subject to automated safety review and protected under Anthropic's information security program.
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 11, 2026 · 08:34 UTC. How this works →