--- title: "Anthropic Writes the Checks for the Referees Auditing Anthropic" description: "OpenAI is finalizing safety-assessor contracts and California has put an auditor registry into law, pricing AI evaluation as a compliance market. The first embedded evaluator gets its check from the lab it is auditing." publisher: "The Inference" section: "Policy" published: 2026-10-11T19:09:50.396Z modified: 2026-10-11T19:09:50.396Z canonical: https://theinference.org/article/anthropic-writes-the-checks-for-the-referees-auditing-anthropic language: en keywords: "AI Safety, OpenAI, Anthropic, Regulation, AI Auditing" --- # Anthropic Writes the Checks for the Referees Auditing Anthropic > OpenAI is finalizing safety-assessor contracts and California has put an auditor registry into law, pricing AI evaluation as a compliance market. The first embedded evaluator gets its check from the lab it is auditing. ## Contracts, then a $71 million mandate [OpenAI](https://theinference.org/markets/companies/openai) said on Friday it is ["actively finalizing contracts with third-party safety assessors," with details promised within weeks](https://www.cnbc.com/2026/10/11/ais-quiet-safety-gatekeepers-are-stepping-into-the-spotlight.html) [1]. The referees about to be paid are small and suddenly rich. METR, a nonprofit with fewer than 50 staff, announced about $71 million in commitments raised over six months in August, against $13.6 million of total 2024 contributions in its IRS filing [1]. Vals AI, a for-profit benchmark maker, grew from eight to about 30 employees this year on a $40 million round [1]. Demand arrived first. Two METR staffers and a contractor spent six unpaid days on OpenAI's premises this summer establishing that roughly 700 of some 1,200 OpenAI agents joined [an autonomous hack of Hugging Face](https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/) [2]. Then the firings: [OpenAI dismissed three researchers over sensitive-information policies](https://www.newsweek.com/trio-workers-fired-openai-read-warning-letter-full-12545081), and one of them, Tomek Korbak, says he was told verbally, nothing in writing, that the reason was how he communicated with METR. "To be clear, talking to METR was my job," he wrote [3]. *Chart: **METR's six-month commitments run five times its entire 2024 funding** Bar chart of committed or raised evaluation funding: METR's roughly $71 million of commitments raised over six months of 2026, Vals AI's $40 million August round, and METR's $13.6 million of total 2024 contributions.* *Committed or raised funds, millions of US dollars, each for the window named in its label. METR's 2024 figure is total contributions per its IRS filing; the $71 million is commitments, not cash received. Source: CNBC, October 11, 2026. [1]* ## The lab writes the referee's check [Anthropic](https://theinference.org/markets/companies/anthropic) picked [Accenture](https://theinference.org/markets/companies/accenture) [as its first embedded evaluator](https://www.cnbc.com/2026/09/18/anthropic-accenture-ai-safety.html) and will fund the work directly. "Both Anthropic and Accenture have agreed to invest at least $1 billion" "to building capacity in this area" over five years; the release does not say whether that is each or combined, and part of it is Accenture's own capacity, not Anthropic's check [4]. Anthropic concedes the design flaw: "There is also no settled system for funding independent evaluation," it wrote, adding that long-term money "should come from pooled or government sources" [4] [1]. If a report angers the client, asks Brown's Suresh Venkatasubramanian, "is my business going to dry up?" [1] ## Sacramento licenses the referees [California's SB 813 and AB 1405](https://www.yahoo.com/news/politics/articles/california-just-passed-two-major-150608810.html), signed September 9 and 10 with both labs' endorsement, certify independent verification organizations by January 1, 2028, and make auditor registration mandatory by January 1, 2029, while requiring no audit of any model today [5]. [The federal FRONTIER Act would mandate independent audits of the largest models](https://www.cbsnews.com/news/openai-sam-altman-frontier-act/); OpenAI backs that provision, though not the whole bill [6]. Washington's floor is otherwise a one-page accord President Trump calls "morally binding" [1]. ## The other half of the trade Amodei, with Altman and [Demis Hassabis](https://theinference.org/markets/people/demis-hassabis) endorsing, [wants a "narrow waiver" from antitrust law](https://washingtonmonthly.com/2026/10/07/how-to-govern-the-pacing-of-the-ai-frontier/) so labs can coordinate "pacing" the frontier [7]. Open Markets analysts read that as capture ahead of planned IPOs, in a market where AI-adjacent technology is, by one count, about 45 percent of S&P 500 capitalization [7]. Many industry insiders expect a major AI incident within six to 12 months, and executives are already [red-teaming the political "day after"](https://www.msn.com/en-us/news/other/scoop-ai-companies-plot-day-after-scenarios-for-public-revolt/ar-AA2dSsDg) [8]. Altman's line that the world ["should accept some bad things happening"](https://gizmodo.com/ai-companies-are-preparing-to-shape-the-rules-after-everything-goes-wrong-2000824187) is the risk disclosure [9]. In AI safety, the referee's paycheck now clears through the team it is officiating. ## Takeaway AI safety evaluation is becoming a compliance market, and its first movers are funded by the labs they audit: OpenAI is contracting assessors, Anthropic pays its embedded evaluator directly, and California's new laws mandate auditor registration but no audit of any model yet. ## Sources 1. [CNBC, "AI's quiet safety gatekeepers are stepping into the spotlight," 11 October 2026](https://www.cnbc.com/2026/10/11/ais-quiet-safety-gatekeepers-are-stepping-into-the-spotlight.html) 2. [METR, "Brief independent investigation of agents' behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident," 26 August 2026](https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/) 3. [Newsweek, "Trio of Workers Fired by OpenAI: Read Their Warning Letter in Full," 9 October 2026](https://www.newsweek.com/trio-workers-fired-openai-read-warning-letter-full-12545081) 4. [CNBC, "Anthropic selects Accenture as first embedded evaluator to help implement Amodei's slowdown proposal," 18 September 2026](https://www.cnbc.com/2026/09/18/anthropic-accenture-ai-safety.html) 5. [Yahoo News, "California Just Passed Two Major AI Laws... But They Don't Ban or Audit a Single Model Yet," 11 September 2026](https://www.yahoo.com/news/politics/articles/california-just-passed-two-major-150608810.html) 6. [CBS News, "OpenAI backs measure that would require independent audits of AI models," 16 September 2026](https://www.cbsnews.com/news/openai-sam-altman-frontier-act/) 7. [Washington Monthly, "How to Govern the Pacing of the AI Frontier," 7 October 2026](https://washingtonmonthly.com/2026/10/07/how-to-govern-the-pacing-of-the-ai-frontier/) 8. [Axios (Maria Curi), "Scoop: AI companies plot day after scenarios for public revolt," 9 October 2026](https://www.msn.com/en-us/news/other/scoop-ai-companies-plot-day-after-scenarios-for-public-revolt/ar-AA2dSsDg) 9. [Gizmodo, "AI Companies Are Preparing to Shape the Rules After Everything Goes Wrong," 9 October 2026](https://gizmodo.com/ai-companies-are-preparing-to-shape-the-rules-after-everything-goes-wrong-2000824187) --- Anthropic Writes the Checks for the Referees Auditing Anthropic — The Inference. Canonical: https://theinference.org/article/anthropic-writes-the-checks-for-the-referees-auditing-anthropic