US CAISI
CAISI (NIST). What it was engaged to test, on what terms, what it found in the document's own words, and what happened afterwards. Every line is a record; every record links to the document it was read from.
- Founded2023
- HQGaithersburg
- Registry idnone (public body)
- TestsPre-deployment testing under model-access agreements.
- Money, as the roster puts itUS government.
- Filesdossier · observatory, sliced to it · graph
- Labs tested
- OpenAI Anthropic Google DeepMind Microsoft
- Span of engagements
- 29 August 2024 – 7 October 2026
- Terms and findings
- 4 of 11 documents state a term · 1 quote a finding
- Money in, US dollars, stated single-recipient rows
- no stated figure 0 in · 0 out
- Seats, by side
- government 2 between 1 3 people: Conrad Stosz, J.J. Allaire, Paul Christiano
- Afterwards
- acted 1 staff move 1
Each count is the number of store records that name this evaluator. An engagement is one lab document naming it for one model; an outcome is an event on the record afterwards; the money, seats and moves are the observatory's rows filtered to it. A cell that cannot be filled says not stated, and each table ends with how many of its rows state each field.
Filters combine: any of the values ticked within a dimension, all of the dimensions at once. A table that has no column for a dimension you have set says so instead of showing rows that cannot match. The address bar carries the slice.
What it was engaged to test 11 engagements, OpenAI, Anthropic, Google DeepMind, Microsoft
In date order, each from the lab document that names the evaluator. Under each: the four terms that document states (access, model stage, publication, payment) and the finding in its words. A term the document is silent on reads not stated; the tags beside the terms are read off their words and are what the console filters on.
- Document: NIST, agreements with Anthropic and OpenAI model access since 2025 [testing agreement: pre- and post-release model access]
- Access window"Each company's Memorandum of Understanding establishes the framework for the U.S. AI Safety Institute to receive access to major new models from each company prior to and following their public release"
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
- Document: NIST, agreements with Anthropic and OpenAI testing agreement: pre- and post-release model access
- Access window"Each company's Memorandum of Understanding establishes the framework for the U.S. AI Safety Institute to receive access to major new models from each company prior to and following their public release"
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
Afterwards: acted staff move - Document: Claude Opus 4 and Sonnet 4 system card
- Access windownot stated
- Model stage"joint pre-deployment testing of the new Claude Opus 4 model was conducted by the US AI Safety Institute (US AISI) and the UK AI Security Institute (UK AISI)"
- Publication"These organizations will also receive a minimally redacted copy of the capabilities report"
- Paymentnot stated
Finding: not stated in the document
- Document: GPT-5.5 system card
- Access windownot stated
- Model stage"CAISI was provided access to a representative launch checkpoint and a checkpoint with reduced refusals for biological capabilities testing"
- Publicationnot stated
- Paymentnot stated
Finding, in the document's words"CAISI's testing did not indicate a broad increase in national security-relevant biological capabilities relative to the GPT-5 helpful-only model"
Afterwards: acted staff moveSources: GPT-5.5 system card 63740c1c85 - Document: Claude Opus 5 system card
- Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
Sources: Claude Opus 5 system card 58e7ec87cf - Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
- Document: Responsible AI Transparency Report 2026 MoU and CRADA with NIST
- Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
- Document: Claude Opus 5.5 system card
- Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
Sources: Claude Opus 5.5 system card 66d824d266 - Document: Claude Sonnet 5.5 system card
- Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
Sources: Claude Sonnet 5.5 system card 1f659de7f9 - Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
- Document: Claude Haiku 5.5 system card
- Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
Sources: Claude Haiku 5.5 system card d6c0bc540d
Dated 11 of 11Model named 8 of 11Terms stated 4 of 11Finding quoted 1 of 11
The terms, as the documents state them
The same engagements as a grid, one row per document and model, so the silences line up: a document that gives an access window but no payment term, or a finding but no publication term, reads that way here.
| Document | Model | Access window | Model stage | Publication | Payment | Finding | Record |
|---|---|---|---|---|---|---|---|
| NIST, agreements with Anthropic and OpenAIAnthropic · 29 August 2024 | no model named | "Each company's Memorandum of Understanding establishes the framework for the U.S. AI Safety Institute to receive access to major new models from each company prior to and following their public release" | not stated | not stated | not stated | not stated | 1755c39447 |
| NIST, agreements with Anthropic and OpenAIOpenAI · 29 August 2024 | no model named | "Each company's Memorandum of Understanding establishes the framework for the U.S. AI Safety Institute to receive access to major new models from each company prior to and following their public release" | not stated | not stated | not stated | not stated | aef969d687 |
| Claude Opus 4 and Sonnet 4 system cardAnthropic · May 2025 | Claude Opus 4 | not stated | "joint pre-deployment testing of the new Claude Opus 4 model was conducted by the US AI Safety Institute (US AISI) and the UK AI Security Institute (UK AISI)" | "These organizations will also receive a minimally redacted copy of the capabilities report" | not stated | not stated | e55f80fc65 |
| GPT-5.5 system cardOpenAI · 23 April 2026 | GPT-5.5 | not stated | "CAISI was provided access to a representative launch checkpoint and a checkpoint with reduced refusals for biological capabilities testing" | not stated | not stated | quoted | 63740c1c85 |
| Claude Opus 5 system cardAnthropic · 24 July 2026 | Opus 5 | not stated | not stated | not stated | not stated | not stated | 58e7ec87cf |
| Claude Fable 5.1 and Mythos 5.1 system cardAnthropic · 1 September 2026 | Fable 5.1 / Mythos 5.1 | not stated | not stated | not stated | not stated | not stated | ddda7a2942 |
| Responsible AI Transparency Report 2026Microsoft · 1 September 2026 | no model named | not stated | not stated | not stated | not stated | not stated | a924e0af9e |
| Claude Opus 5.5 system cardAnthropic · 22 September 2026 | Opus 5.5 | not stated | not stated | not stated | not stated | not stated | 66d824d266 |
| Claude Sonnet 5.5 system cardAnthropic · 28 September 2026 | Sonnet 5.5 | not stated | not stated | not stated | not stated | not stated | 1f659de7f9 |
| Gemini 4 Argon Safety and Security ReportGoogle DeepMind · October 2026 | Gemini 4 Argon | not stated | not stated | not stated | not stated | not stated | be7801e2b6 |
| Claude Haiku 5.5 system cardAnthropic · 7 October 2026 | Haiku 5.5 | not stated | not stated | not stated | not stated | not stated | d6c0bc540d |
Access window 2 of 11Model stage 2 of 11Publication 1 of 11Payment 0 of 11Finding, quoted 1 of 11
What happened afterwards 2 on record
Events a party published after an engagement: an incident on a model this evaluator's engagement covered, a correction or added disclosure, a later finding that cut against an earlier one, a lab's statement that it acted or did not act on the finding, or a move from this evaluator to a lab it graded. A row the source attaches to this evaluator by name is marked named; one that reaches it through the model it tested is marked via the model. The kind is what the record states, not why.
| Kind | Developer | Model | What the record states | When | Source | Id |
|---|---|---|---|---|---|---|
| actednamed | OpenAI | no model named | The lab's account of the institute's product red-teaming: CAISI "identified two novel security vulnerabilities in ChatGPT Agent"; "These attacks were immediately reported to OpenAI, and fixed by OpenAI within one business day"Engagements: 63740c1c85, aef969d687 | 12 September 2025 | OpenAI, CAISI and UK AISI update | 8fe6218c20 |
| staff movenamed | OpenAI | no model named | Paul Christiano: ARC; US AI Safety Institute (Apr 2024); CAISI senior technical adviser; back to ARC as executive director (Aug 2026) -> OpenAI Foundation board (9 Sep 2026)Engagements: 63740c1c85, aef969d687 | 9 Sep 2026 | NIST; OpenAI Foundation; ARC | 47b741082c |
Dated 2 of 2Model named 0 of 2Tied to an engagement 2 of 2Evaluator named by the record 2 of 2
The money on record 0 rows
Rows of the observatory's money table that name this evaluator as recipient (in) or as funder (out), each as the record states it: a grant table line, a contract notice, a Form 990 schedule or the party's own disclosure. A figure quoted in prose is marked and never summed.
| Direction | Funder | Recipient | Amount, as recorded | When | Record | Source | Id |
|---|---|---|---|---|---|---|---|
| no money row names this evaluator | |||||||
Amount stated 0 of 0Dated 0 of 0Purpose stated 0 of 0Funder profiled 0 of 0Recipient on the roster 0 of 0
Seats and moves 3 seats · 3 moves
Seats read off a team page, a governance page, an appointment notice or a filing that name this organization, and the dated moves into or out of it, from the observatory's own tables.
| Person | Side | Seat, as the record states it | Role | Dates | Source | Id |
|---|---|---|---|---|---|---|
| Paul Christianoslice | government | Head of AI Safety, US AI Safety Institute, from Apr 2024 | Head of AI Safety | Apr 2024 – current, as read | ARC; METR, 19 Sep 2023; OpenAI Foundation; Anthropic | 193ffc692c |
| Paul Christianoslice | government | Senior Technical Advisor, CAISI, recused from OpenAI matters and model evaluations | Senior Technical Advisor | dates not stated | ARC; METR, 19 Sep 2023; OpenAI Foundation; Anthropic | 7103ade165 |
| Conrad Stoszslice | between | Head of Governance, Transluce, from 22 Oct 2025; "previously led the U.S. Center for Standards and Innovation"names this evaluator alongside its own organization | Head of Governance | 22 Oct 2025 – current, as read | Transluce | f2aada7bba |
Role read 3 of 3Organization read 3 of 3Organization profiled 3 of 3Dated 2 of 3
Moves
| Person | From | To | When | Source | Id | |
|---|---|---|---|---|---|---|
| Paul Christianoslice | OpenAI (2017 to 2021) | → | ARC; US AI Safety Institute (Apr 2024); CAISI senior technical adviser; back to ARC as executive director (Aug 2026); OpenAI Foundation board (9 Sep 2026) | 2021 to Sep 2026 | NIST; OpenAI Foundation; ARC | 30fdcfdee3 |
| Conrad Stoszslice | US Center for AI Standards and Innovation (led it) | → | Transluce (Head of Governance) | 22 Oct 2025 | Transluce | 04210710e1 |
| J.J. Allaireslice | UK AI Security Institute (technical staff, Nov 2024) | → | Meridian Labs (co-founder); visiting researcher at US CAISI | 2025 | UK AISI; Meridian Labs | d61b436c68 |
From profiled 3 of 3To profiled 3 of 3Person profiled 3 of 3Dated 3 of 3