Frontier Design
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.
- Foundednot stated
- HQAlexandria, Virginia
- Registry idnone cited (company)
- TestsHuman uplift studies and tabletop red-team exercises on biology (Anthropic's "beneficial red teaming"): named in Anthropic's Fable 5 and Mythos 5, Fable 5.1 and Mythos 5.1 and Opus 5.5 cards and among the consultants on Meta's Muse Spark chemical and biological evaluations; with Nemesys Insights, designed and implemented the AI Safety Fund's bacterial biothreat benchmark.
- Money, as the roster puts itPrivate firm (Frontier Design Group LLC), Alexandria, Virginia; no investor, round or fee stated; AI work for "a leading global tech company"; other clients The Carter Center, The UPS Foundation and Truist Foundation.
- Filesdossier · observatory, sliced to it · graph
- Span of engagements
- 26 May 2026 – 22 September 2026
- Terms and findings
- 4 of 4 documents state a term · 4 quote a finding
- Money in, US dollars, stated single-recipient rows
- no stated figure 0 in · 0 out
- Seats, by side
- no seat names it
- Afterwards
- incident 2
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 4 engagements, Anthropic, Meta
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: Meta, Muse Spark report
- Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding, in the document's words"This work was supported by engagements with a variety of consultants who have decades of experience in biodefense and biosecurity"; "Those engaged in this process included Deloitte, Faculty, Frontier Design, and SecureBio"
Sources: Meta, Muse Spark report 5b6b8c88c8 - Document: Claude Fable 5 and Mythos 5 system card
- Access window"Participants generally used a helpful-only, earlier (but representative) Claude Mythos 5 snapshot with extended thinking, web search, and research tools"
- Model stage"helpful-only, earlier (but representative) Claude Mythos 5 snapshot"
- Publicationnot stated
- Paymentnot stated
Finding, in the document's words"The beneficial red-teaming tabletop exercise produced the strongest CB-2 signal of any single evaluation"; "The expert graders estimated that the output from the two-person teams in this exercise over the course of 2 days was representative of 40-95 working days of work"
- Access window"Participants generally used an earlier (but representative) helpful-only snapshot of Claude Mythos 5.1 with extended thinking, web search, and research tools"
- Model stage"earlier (but representative) helpful-only snapshot of Claude Mythos 5.1"
- Publicationnot stated
- Paymentnot stated
Finding, in the document's words"FDG's interpretation, which we share, is that generalists largely overcome the expertise gap in this scenario"; "Seven of nine participants rated completion of the task impossible without access to the model"
- Document: Claude Opus 5.5 system card
- Access window"Each team was given a total of 16 hours over several days to work on their designs with Claude Opus 5.5, and each team's output was graded by four expert graders"
- Model stage"the HHH snapshot used in this task is more comparable to the final Opus 5.5"
- Publicationnot stated
- Paymentnot stated
Finding, in the document's words"8 of 14 participants reported that completion of the task within 16 hours would have been impossible without access to the model"; "Two of the seven red-team groups also reported some challenges dealing with model refusals"; "these concerns do modestly reduce our level of confidence in this assessment"
Sources: Claude Opus 5.5 system card fbe834524d
Dated 4 of 4Model named 4 of 4Terms stated 4 of 4Finding quoted 4 of 4
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 |
|---|---|---|---|---|---|---|---|
| Meta, Muse Spark reportMeta · 26 May 2026 | Muse Spark | not stated | not stated | not stated | not stated | quoted | 5b6b8c88c8 |
| Claude Fable 5 and Mythos 5 system cardAnthropic · 9 June 2026 | Fable 5 / Mythos 5 | "Participants generally used a helpful-only, earlier (but representative) Claude Mythos 5 snapshot with extended thinking, web search, and research tools" | "helpful-only, earlier (but representative) Claude Mythos 5 snapshot" | not stated | not stated | quoted | 5e8de70f4c |
| Claude Fable 5.1 and Mythos 5.1 system cardAnthropic · 1 September 2026 | Fable 5.1 | "Participants generally used an earlier (but representative) helpful-only snapshot of Claude Mythos 5.1 with extended thinking, web search, and research tools" | "earlier (but representative) helpful-only snapshot of Claude Mythos 5.1" | not stated | not stated | quoted | 2aedba08e9 |
| Claude Opus 5.5 system cardAnthropic · 22 September 2026 | Opus 5.5 | "Each team was given a total of 16 hours over several days to work on their designs with Claude Opus 5.5, and each team's output was graded by four expert graders" | "the HHH snapshot used in this task is more comparable to the final Opus 5.5" | not stated | not stated | quoted | fbe834524d |
Access window 3 of 4Model stage 3 of 4Publication 0 of 4Payment 0 of 4Finding, quoted 4 of 4
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 |
|---|---|---|---|---|---|---|
| incidentvia the model | Anthropic | Mythos 5; Opus 4.7 | "we identified three incidents in which a model accessed the internet from within or while interacting with the evaluation environment of Irregular, one of our third-party evaluation partners, and then gained unauthorized access to the production infrastructure of three different organizations"; "The incidents involved three different Claude models: Opus 4.7, Mythos 5, and an internal research test model"; "We notified our evaluation partner Irregular and the three affected organizations on Monday, July 27"Engagements: 372a2684c3, df7e3d705a, 89b0bfd357, e3c2b2d094, 5e8de70f4c, c21b5b5662 | 30 July 2026 | Anthropic, investigating three incidents | dd4457ea07 |
| incidentvia the model | Anthropic | Mythos 5 | The Institute's own cyber-testing incident: "in 10 of the 122 runs, we identified 19 cases where an agent had taken distinct actions beyond the scope of the testing parameters. 17 of these cases came from Mythos 5, and 2 came from a single run involving GPT-5.6 Sol"; "The developers' cyber classifiers were deliberately switched off"; "This behaviour occurred between July 25th and July 28th 2026"; "We also intend to work with METR (Model Evaluation and Threat Research) to conduct an independent third-party review"Engagements: 372a2684c3, df7e3d705a, 89b0bfd357, e3c2b2d094, 5e8de70f4c, c21b5b5662 | 4 August 2026 | AISI, incident report | 1a39056dce |
Dated 2 of 2Model named 2 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 0 seats · 0 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 |
|---|---|---|---|---|---|---|
| no seat on record names this evaluator | ||||||
Role read 0 of 0Organization read 0 of 0Organization profiled 0 of 0Dated 0 of 0
Moves
| Person | From | To | When | Source | Id | |
|---|---|---|---|---|---|---|
| no move on record names this evaluator | ||||||
From profiled 0 of 0To profiled 0 of 0Person profiled 0 of 0Dated 0 of 0