System Card: Claude Haiku 5.5
Anthropic's system card for Claude Haiku 5.5, dated October 7, 2026 on its cover.
- Issued by
- Anthropic
- Authors credited
- No authorship claim recorded yet. The source record lists Anthropic.
- Dates
- Published 2026-10-07 issue evidence
The date printed on the cover, day precision; the card does not say whether it is an issue or publication date. No changelog or revision date was found.
- Exact source
- System Card: Claude Haiku 5.5 Anthropic · published 2026-10-07 · Retrieved for the candidate pack on 2026-11-02 (SHA-256 d5dbaab6996c02ad6ff28b032ce5a9c7582e59af09b5d1bc0975cf0235535da9) directly from the publisher’s host; the request redirected to https://assets.claude.com/e1080d6bf5ae2018ea3c2f414064be03232f5be5.pdf. Retrieved again on 2026-11-03 with the same SHA-256. The archive capture was found by an index search of the Internet Archive on 2026-11-03; it was not opened, and its bytes are not authenticated.Archived copy, captured 2026-10-26: System Card: Claude Haiku 5.5. Captured 19 days after the recorded publication date. An archive copy shows only what the archive retrieved on 2026-10-26. It does not prove what the page said earlier, and it can differ from the copy checked on 2026-11-02.
Appears in
No collection holds a record about this document yet.
Versions
anthropic-claude-haiku-5-5-system-card: 1 recorded version, oldest first, published by Anthropic. A version is shown as replacing another only where a record says so. Date order alone never implies it.
- 2026-10-07 · publishedAs dated October 7, 2026 on its cover (this version)
This version
Provisions in this version · 14 quoted · 2 not found
A limited selection of quoted provisions: what this edition says about thresholds, required actions, who decides and exceptions, and other statements of the text. It is not a full summary and says nothing about implementation. A “not found” record states which part of the edition was inspected. It is not a claim that the text was deleted.
- Model-card questionCard date“System Card: Claude Haiku 5.5 October 7, 2026”Cover, PDF p. 1Inspect claim & evidence ↗
- Model-card questionRelease date“Claude Haiku 5.5 is a new large language model from Anthropic, released for general access.”Section 1 'Introduction', PDF p. 8Inspect claim & evidence ↗
- Model-card questionFramework cited“we describe our assessment of Claude Haiku 5.5 for catastrophic risks relevant to our Responsible Scaling Policy (RSP)”Section 2.1 'Introduction', PDF p. 11Inspect claim & evidence ↗
- Model-card questionLink to framework“we describe our assessment of Claude Haiku 5.5 for catastrophic risks relevant to our Responsible Scaling Policy (RSP)”Section 2.1 'Introduction', PDF p. 11 (hyperlinked words)Inspect claim & evidence ↗
- Model-card questionFramework determination“As such, we determine that Haiku 5.5 does not cross our CB-2 or Autonomy-2 RSP thresholds.”Section 2.1 'Introduction', PDF p. 11Inspect claim & evidence ↗
- Model-card questionSafeguards tied to determination“However, we do treat Haiku 5.5 as meeting our CB-1 and Autonomy-1 thresholds, and we apply the corresponding mitigations from our RSP.”Section 2.1 'Introduction', PDF p. 11Inspect claim & evidence ↗
- Model-card questionChemical and biological“As such, we determine that Haiku 5.5 does not cross our CB-2 or Autonomy-2 RSP thresholds.”Section 2.1 'Introduction', PDF p. 11Inspect claim & evidence ↗
- Model-card questionCyber“Claude Haiku 5.5 is not a frontier model, but it is a relatively significant step up in cyber capabilities over Haiku 4.5.”Section 3.1 'Introduction', PDF p. 23Inspect claim & evidence ↗
- Model-card questionAutonomy and AI R&D“Autonomy threat model 2 is not applicable to Haiku 5.5.”Section 2.3.1.1 'How Haiku 5.5 affects or changes analysis from our most recent Risk Report', PDF p. 19Inspect claim & evidence ↗
- Model-card questionAlignment and deception“our overall assessment is that the risk of catastrophic harm caused by misalignment of Claude Haiku 5.5 is also low.”Section 2.4.2 'Updated overall assessment of alignment risk', PDF p. 22Inspect claim & evidence ↗
- Model-card questionRed-teaming“We use Shade, an external adaptive red-teaming tool from Gray Swan, to evaluate our models’ robustness to prompt injection in coding environments.”Section 5.2.2.1 'Coding', PDF p. 51Inspect claim & evidence ↗
- Model-card questionExternal testers“Gray Swan trains the attacker on the test scenarios against earlier Claude models, then runs it unchanged against the latest models on the same scenarios.”Section 5.2.2.1 'Coding', PDF p. 51Inspect claim & evidence ↗
- Model-card questionRelease decision“As such, we are releasing Haiku 5.5 with the same safeguards we deployed for Sonnet 5 and Opus 5”Section 2.2.4 'Conclusions', PDF p. 18Inspect claim & evidence ↗
- Model-card questionStated limits“these measures are limited: past assessments have not fully captured the potential severity of misbehavior our models were capable of”Section 6.1.2 'Key findings on safety and alignment', PDF p. 56Inspect claim & evidence ↗
- Not found in the inspected textPersuasionInspected: System Card: Claude Haiku 5.5: Executive summary, Sections 2, 4.1, 6.1–6.2 and term search across the card.Inspect claim & evidence ↗
- Not found in the inspected textGovernment instituteInspected: System Card: Claude Haiku 5.5: Sections 1.6, 2, 3, 5.2 and term search across the card.Inspect claim & evidence ↗
Earliest version recorded here; no replacement is recorded.
Issue and authorship records
Documentary relations
Each relation is listed separately. Citing, crediting and reporting the use of a method are different acts, and none of them shows agreement or influence by itself.
Not recorded for this document: cites · cited by · explicitly credits · reports using a method from · method reported as used by · applies parts of · parts applied by · discusses · reports evaluation of. A missing record does not show that the relation does not exist.