Epoch AI
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.
- Founded2021
- HQSan Francisco
- Registry idEIN 99-4050541
- TestsBenchmarks and trends (FrontierMath).
- Money, as the roster puts itOpen Philanthropy; OpenAI funded FrontierMath; Good Ventures $70,000 for FrontierMath improvements, tax year to 30 Jun 2025.
- Filesdossier · observatory, sliced to it · graph
- Models
- FrontierMath
- Span of engagements
- 23 January 2025 – 7 April 2026
- Terms and findings
- 1 of 2 documents state a term · 0 quote a finding
- Money in, US dollars, stated single-recipient rows
- $95,600,897 24 in · 0 outfrom Coefficient Giving · European AI Office · Good Ventures Foundation · Jaan Tallinn · METR · Nat Friedman · OpenAI · Schmidt Sciences · Sentinel Bio · Survival and Flourishing Fund · UK AI Security Institute
- Seats, by side
- evaluator 5 between 1 6 people: Ajeya Cotra, Ege Erdil, Jaime Sevilla, Marius Hobbhahn, Miles Brundage, Tamay Besiroglu
- Afterwards
- correction 2 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 2 engagements, OpenAI, Anthropic
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: Epoch AI, the FrontierMath commission
- Access window"OpenAI retains ownership of these questions and has access to the problems and solutions, with the exception of a holdout set"
- Model stagenot stated
- Publication"Epoch AI is free to conduct and publish evaluations of any models using the FrontierMath problem set commissioned by OpenAI. However, we cannot share the questions and answers with other parties without written permission from OpenAI"
- Payment"OpenAI commissioned Epoch AI to produce 300 advanced math problems for AI evaluation that form the core of the FrontierMath benchmark"
Finding: not stated in the document
Afterwards: correction - Document: Anthropic, Claude Mythos Preview system card, Apr 2026 Its capabilities index forked by Anthropic: "we fork from Epoch AI's implementation of this work, the Epoch Capabilities Index (ECI)"
- Access windownot stated
- Model stagenot stated
- Publicationnot stated
- Paymentnot stated
Finding: not stated in the document
Dated 2 of 2Model named 1 of 2Terms stated 1 of 2Finding quoted 0 of 2
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 |
|---|---|---|---|---|---|---|---|
| Epoch AI, the FrontierMath commissionOpenAI · 23 January 2025 | FrontierMath | "OpenAI retains ownership of these questions and has access to the problems and solutions, with the exception of a holdout set" | not stated | "Epoch AI is free to conduct and publish evaluations of any models using the FrontierMath problem set commissioned by OpenAI. However, we cannot share the questions and answers with other parties without written permission from OpenAI" | "OpenAI commissioned Epoch AI to produce 300 advanced math problems for AI evaluation that form the core of the FrontierMath benchmark" | not stated | e53c356823 |
| Anthropic, Claude Mythos Preview system card, Apr 2026Anthropic · 7 April 2026 | no model named | not stated | not stated | not stated | not stated | not stated | 15e43e308c |
Access window 1 of 2Model stage 0 of 2Publication 1 of 2Payment 1 of 2Finding, quoted 0 of 2
What happened afterwards 3 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 |
|---|---|---|---|---|---|---|
| correctionnamed | OpenAI | FrontierMath | "On 2026-06-12, we released a major update, addressing errors in 42% of problems"; "We released v2, which corrected 123 problems in Tiers 1-3 and 12 problems in Tier 4. We also removed 5 problems from Tiers 1-3 and 7 problems from Tier 4"Engagement: e53c356823 | 12 June 2026 | Epoch AI, FrontierMath Tier 4 (v2) | f2ba88ddba |
| staff movenamed | Google DeepMind | no model named | Tamay Besiroglu: Epoch AI (co-founder, associate director); Mechanize (co-founder, Apr 2025) -> Google DeepMind (research scientist, by Sep 2026) | Sep 2026 | Epoch AI; Mechanize; his post | 75f848d85a |
| correctionnamed | OpenAI | GPT-6 Astra | Its chief executive on the model's Epoch Capabilities Index score: "An update on Astra's ECI score. This is a major step down from the original score we reported (but still within the uncertainty range we expected). It is explained by lower performance on long-horizon software engineering tasks" | 16 September 2026 | Jaime Sevilla on X | 5d33e90812 |
Dated 3 of 3Model named 2 of 3Tied to an engagement 1 of 3Evaluator named by the record 3 of 3
The money on record 24 rows · $95,600,897 in, stated single-recipient 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 |
|---|---|---|---|---|---|---|---|
| in | Coefficient Giving | Epoch AI | $25,113,611Eight grants totalling $25,113,611 (Jun 2022 to Jul 2026), including $8,500,000 (Jun 2025)including $8,500,000 | Jun 2022 | party's own page | Epoch AI; Coefficient Giving | f97c5b81cf |
| in | Coefficient Giving | Epoch AI | $68,796,286$68,796,286 (Aug 2026), "a $68 million renewal over 3 years""a $68 million renewal over 3 years" | Aug 2026 | party's own page | Epoch AI; Coefficient Giving | 2f08995fb2 |
| in | Survival and Flourishing Fund; Jaan Tallinn | Epoch AI | $195,000$195,000 (May 2026) | May 2026 | party's own page | Epoch AI; SFF-2026 | bb9e1db106 |
| in | Jaan Tallinn | Epoch AI | $600,000$600,000 (Jan 2025) | Jan 2025 | party's own page | Epoch AI; SFF-2026 | e6b88d3b90 |
| in | Jaan Tallinn | Epoch AI | $646,000SFF-2026 recommends $646,000 in all (Moskovitz $542,000 of it)SFF- recommends | 2026 | party's own page | Epoch AI; SFF-2026 | aafd049fe8 |
| in | Schmidt Sciences | Epoch AI | no amount publishedNamed donor (Dec 2025), no amount | Dec 2025 | party's own page | Epoch AI | 08a0bf8e3c |
| in | OpenAI | Epoch AI | no amount publishedCommissioned and funded FrontierMath's 300-problem core | not dated | party's own page | Epoch AI; Sevilla | 2555c5299b |
| in | OpenAI | Epoch AI | no amount published"OpenAI purchased access to the verifiers" (Apr 2026) | Apr 2026 | party's own page | Epoch AI; Sevilla | e85d2d0a4a |
| in | UK AI Security Institute | Epoch AI | no amount publishedEpoch lists UK DSIT as a paying partner | not dated | party's own page | METR; Epoch AI; Manifund; Generality Labs | f46738332c |
| in | European AI Office | Epoch AI | no amount publishedEpoch: a paying partner (2026) | 2026 | party's own page | METR; Epoch AI; Transluce; FAR.AI | f59b15efd6 |
| in | METR | Epoch AI | no amount publishedA 2025 software-engineering benchmark engagement, funded by METR | 2025 | party's own page | Epoch AI | 7c3cc5bec8 |
| in | Nat Friedman | Epoch AI; MECHANIZE; Artificial Analysis | no amount publishedFunded a pilot of FrontierMath | not dated | party's own page | Epoch AI; Mechanize; Artificial Analysis; AIUC, Series A; Fortune, Oct 8 2025; Meta memo via CNBC | 16a5ebe3a8 |
| in | Nat Friedman | Epoch AI; MECHANIZE; Artificial Analysis | no amount publishedfirst-named backer of Mechanize | not dated | party's own page | Epoch AI; Mechanize; Artificial Analysis; AIUC, Series A; Fortune, Oct 8 2025; Meta memo via CNBC | cf9b872cbf |
| in | Nat Friedman | Epoch AI; MECHANIZE; Artificial Analysis | no amount publishedfirst-named of the eight backers on Artificial Analysis's About page, no amount stated | not dated | party's own page | Epoch AI; Mechanize; Artificial Analysis; AIUC, Series A; Fortune, Oct 8 2025; Meta memo via CNBC | 5e85e6667c |
| in | Nat Friedman | Epoch AI; MECHANIZE; Artificial Analysis | $55,000,000named on AIUC's team page among its $55 million investors, and AIUC's Series A post counts "our $15 million seed round from NFDG"investors, and AIUC's Series A post counts "our $15 million seed round from NFDG" | not dated | party's own page | Epoch AI; Mechanize; Artificial Analysis; AIUC, Series A; Fortune, Oct 8 2025; Meta memo via CNBC | 4aef243df2 |
| in | Nat Friedman | Epoch AI; MECHANIZE; Artificial Analysis | no amount publishedFortune names him among the seeders of Situational Awareness LP | not dated | party's own page | Epoch AI; Mechanize; Artificial Analysis; AIUC, Series A; Fortune, Oct 8 2025; Meta memo via CNBC | 0291b6507f |
| in | Nat Friedman | Epoch AI; MECHANIZE; Artificial Analysis | no amount publishedsince June 2025 co-lead of Meta Superintelligence Labs for AI products and applied research, per Meta's 30 June 2025 memo | June 2025 | party's own page | Epoch AI; Mechanize; Artificial Analysis; AIUC, Series A; Fortune, Oct 8 2025; Meta memo via CNBC | 835d93c2fa |
| in | Good Ventures Foundation | Epoch AI | $70,000$70,000 (improvements to the FrontierMath benchmark)improvements to the FrontierMath benchmark | not dated | 990-PF | ProPublica, Good Ventures FY2024 Form 990-PF | afb66f19f8 |
| in | Sentinel Bio | Epoch AI | $80,000$80,000 (Aug 2024), | Aug 2024 | party's own page | Sentinel Bio, Research & Projects; Epoch AI; Sentinel Bio, RAND | 0ca4edb7b2 |
| in | Sentinel Bio | Epoch AI | $85,000$85,000 (Sep 2025), | Sep 2025 | party's own page | Sentinel Bio, Research & Projects; Epoch AI; Sentinel Bio, RAND | 4eee235c5d |
| in | Sentinel Bio | Epoch AI | $15,000$15,000 (Mar 2026) on Sentinel's list, and Epoch's own page lists | Mar 2026 | party's own page | Sentinel Bio, Research & Projects; Epoch AI; Sentinel Bio, RAND | cb7f8d0445 |
| in | Sentinel Bio | Epoch AI | restated$85,000 (Sep 2025) | Sep 2025 | party's own page | Sentinel Bio, Research & Projects; Epoch AI; Sentinel Bio, RAND | 5b2443a2b0 |
| in | Sentinel Bio | Epoch AI | restated$80,000 (Sep 2024) | Sep 2024 | party's own page | Sentinel Bio, Research & Projects; Epoch AI; Sentinel Bio, RAND | 17c9992e24 |
| in | European AI Office | METR; Epoch AI | €1,167,484EUR 1,167,484Lot 3, loss of control risk modelling and evaluation | December 2025 | contract notice | TED | 421931534e |
Amount stated 11 of 24Dated 16 of 24Purpose stated 6 of 24Funder profiled 24 of 24Recipient on the roster 24 of 24
Seats and moves 6 seats · 4 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 |
|---|---|---|---|---|---|---|
| Ajeya Cotraslice | evaluator | Board of Directors, Epoch AI (from its early-2025 spin-out) | Board of Directors | 2025 – current, as read | Epoch AI | 8f41b9b17e |
| Marius Hobbhahnslice | evaluator | part-time research fellow, Epoch, 2022 to 2023 | part-time research fellowformer | 2022 – 2023 | Hobbhahn CV | ebf0e84d58 |
| Jaime Sevillaslice | evaluator | Co-founder, CEO and board member, Epoch AI; Treasurer on its FY2024 990-EZ | Co-founder, CEO and board member | dates not stated | Epoch AI; ProPublica; jaimesevilla.me | dcbd014ebf |
| Tamay Besirogluslice | evaluator | Co-founder, Epoch AI; associate director to 2025, then research advisor | Co-founderformer | – 2025 | Epoch AI; Mechanize; his post | 35e2a3cf5e |
| Ege Erdilslice | evaluator | Researcher, Epoch AI, left by Mar 2025 | Researcherformer | – Mar 2025 | Epoch AI; Mechanize | 12a61628bc |
| Miles Brundageslice | between | Strategy advisor, Epoch AI | Strategy advisor | dates not stated | Epoch AI | 7284f229c3 |
Role read 6 of 6Organization read 6 of 6Organization profiled 6 of 6Dated 4 of 6
Moves
| Person | From | To | When | Source | Id | |
|---|---|---|---|---|---|---|
| Ajeya Cotraslice | Coefficient Giving , technical AI safety grantmaking; Redwood Research board (to Aug 2024) | → | METR technical staff; Epoch AI board | 2024 to 2026 | Coefficient Giving; METR; Epoch AI | 8a94650c89 |
| Tamay Besirogluslice | Epoch AI (co-founder, associate director) | → | Mechanize (co-founder, Apr 2025); Google DeepMind (research scientist, by Sep 2026) | 2025 to 2026 | Epoch AI; Mechanize; his post | d6d4d94fda |
| Ege Erdilslice | Epoch AI | → | Mechanize (co-founder) | Apr 2025 | Mechanize | 8b99131748 |
| Marius Hobbhahnslice | Epoch (part-time research fellow) | → | Apollo Research (co-founder) | 2023 | Hobbhahn CV | af0df61432 |
From profiled 4 of 4To profiled 4 of 4Person profiled 4 of 4Dated 4 of 4
Filings
- IRS (via ProPublica Nonprofit Explorer) 99-4050541: FY2024 Form 990-EZ read 2026-10-17ProPublica Nonprofit Explorer 07d1145c35