RAJIV DATTANI

Behavioral Archetype

THE UNDERWRITER WHO SITS ON THE GRADER’S BOARD — Subject spent ten years in insurance, as a partner at McKinsey and earlier at Apollo, became chief operating officer of the evaluator METR, then co-founded the Artificial Intelligence Underwriting Company (AIUC), a company that writes a standard for AI agents, conducts the testing behind its certificate and tells buyers that insurers offer coverage to the certified. He remains on METR’s board. This profile scores the position, not the man.

The position is the seam between the body that grades models and the body that prices what the grade means. AIUC’s pages say the second job is its business: “AI needs the same combination of standards, testing and insurance,” he told SiliconANGLE. His own posts say how he would wire the two together: “Insure the labs and have the insurer choose & pay evaluators.” The documents also show the seam is disclosed on both sides: METR’s about page lists him as “Advisor and Board Member,” and AIUC’s team page says “Board member @ METR.” What the documents do not show is any engagement between the two organizations.

Essence Indicators

• Co-founder, AIUC. AIUC’s team page lists two co-founders and describes him: “Board member @ METR. Former McKinsey Insurance partner” (AIUC, team). The Lawfare author note: “Rajiv Dattani is a co-founder of the Artificial Intelligence Underwriting Company (AIUC) and a former McKinsey insurance partner” (Lawfare). He is the byline or co-byline on AIUC’s Series A, KPMG, Schellman and MITRE posts (AIUC, Series A; AIUC, KPMG; AIUC, Schellman).
• Advisor and board member, METR. METR’s about page lists “Rajiv Dattani Advisor and Board Member.” His METR team page: “Rajiv is building AI auditing and insurance mechanisms at the Artificial Intelligence Underwriting Company. His previous roles include COO at METR, Founding Advisor at Transluce, and Partner at McKinsey” (METR, About; METR, Rajiv Dattani).
• COO and Treasurer, METR, FY2024. ProPublica’s summary of METR’s Form 990 for the year ending December 2024 (filed 16 November 2025) lists “Rajiv Dattani (Coo And Treasurer),” with reportable compensation of $230,384 and $630 of other compensation, against total revenue of $13,639,155 for the organization (ProPublica, Model Evaluation And Threat Research). Compensation as ProPublica prints it; no compensation for the board seat appears on any page read.
• Earlier career. His X post of 23 September 2026: “I spent 10 years in insurance, at Apollo and as a McKinsey partner” (Dattani, X). SiliconANGLE, as reported: “Dattani spent time as a partner in McKinsey & Co.’s insurance practice, then served as chief operating officer of AI model evaluation nonprofit METR” (SiliconANGLE, 15 September 2026).
• Founding Advisor, Transluce, former. The listing is on METR’s page above. Transluce’s own About page lists no advisers by name.
• Halcyon’s page. Halcyon Futures carries his testimonial under the labels “Career Transition Grant,” “Halcyon Retreat,” “Invested in Every Round” and “$16M raised from top VCs”: “Halcyon has been instrumental for my journey to founding AIUC – If I were starting something new, Mike would be my first call” (Halcyon Futures). The labels are Halcyon’s.
• Listed author, AVERI’s flagship paper. The paper’s author list names “Rajiv Dattani²¹” and the affiliation “AI Underwriting Company”; AVERI’s about page lists “the AI Underwriting Company” first among its funders (AVERI, Frontier AI Auditing; AVERI, About).
• Signatory, AI Evaluator Forum letter. The Forum’s 18 September 2026 letter lists “Rajiv Dattani Co-founder, Artificial Intelligence Underwriting Company (AIUC)” among its 200-plus signatories (AI Evaluator Forum, letter).
• Co-author, AIUC research. “Dr. Keri Pearlson at MIT Sloan and Rajiv Dattani at AIUC have written a paper to AI-proof the Board and the C-Suite” (21 July 2025) (AIUC, AI-Proofing The Board and C-suite); the GovAI-listed report “Underwriting the Agent Economy” (22 July 2026), which he calls “Our new report” (GovAI; Dattani, X).

Social Persona / Impression Management

Immediate impression: An insurance person in an evaluator’s world. His posts carry the historian’s version of the argument: “Insurance gives someone a reason to figure out how a technology actually works. If you’re paying when a boiler explodes, it helps to know why boilers explode” (Dattani, X). The same thread runs Hartford Steam Boiler, Underwriters Laboratories, crash-test ratings and nuclear insurance as precedents.

Energy: Builder register. He announces certifications (“More than 100,000 lawyers use Harvey, now AIUC-1 certified”), posts congratulations to colleagues at METR, AVERI and the UK AI Security Institute, and writes in numbered lists (Dattani, X). On 25 February 2026: “Most certificates are a PDF and a logo. Excited to be building a trust mark that matters” (Dattani, X).

Impression management strategy: Argues the case and then lists its limits. After five replies to common objections to a lab mutual he adds, in his own words, “some important limitations of a mutual insurance model that I think might be underrated,” among them “Labs would need to agree” and “Liability is still murky” (Dattani, X). Replies to critics are short and pointed: “I don’t think mandatory audits solves this problem – we need a 3rd party to select auditors” (Dattani, X).

Forensic Archetype Comparison

Psychometric Assessment

Trait scores are observations of the documented public role, not claims about private character. The record is mostly organizational pages and his own posts, and each evidence cell says which.

Big Five (OCEAN) (inferred from the public role only):

Dark Triad (evidence-bound):

MBTI: ENTP-adjacent, inferred from the documented public role only. The rationale is the work product: an arguer from precedent who builds institutions around the argument and lists its weaknesses in the same post (Dattani, X).

Threat Assessment

Calibrated against Mike McCormick (5/5/4, hazard 46.7), the funder with two evaluator board seats; Marco Mascorro (4/5/3, hazard 40), the investor who advises METR; and Kyle Scott (4/4/3, hazard 36.7), the former METR operations officer. He sits level with McCormick on leverage because each holds board seats alongside a money or standards role; one point below on insulation because his company’s income is fees from the graded and McCormick’s funder’s is donations; and level on durability because the AIUC-1 standard is the artifact and McCormick’s organizations are. Adam Gleave (6/7/6, hazard 55) is above him on every component: he sits on METR’s board too and runs a second evaluator.

Score basis. Atk 5 is leverage: AIUC’s pages say its certificate is what insurers read and its releases name KPMG, Cursor, ElevenLabs and others as certified, and METR lists him as “Advisor and Board Member”; the point under a higher number is that he runs neither organization’s evaluations and nothing obliges a lab or buyer to use the certificate (AIUC, Series A; METR, About). Def 4 is insulation: venture funding from named investors, a customer base that pays for the certification AIUC conducts and issues, no tenure or statutory seat (AIUC, team). Hp 4 is durability: the standard is re-issued quarterly and carried in the CSA STAR registry, and the METR seat is transferable; the points off are the company’s age and the absence of a statute or multi-year contract (AIUC-1, CSA).

Alliance Map

These run at the same time. One person holds a board seat at the grader that OpenAI and Anthropic hand their models to, co-founds the insurer whose essays name that grader as an example contractor, is listed on the paper of an audit advocate that names his company among its funders, was a founding adviser to a third evaluator, and carries credit on the incubator page whose chief executive sits on the boards of two others. METR, AIUC, AVERI, Halcyon and the Forum each disclose their part. The overlap is the finding; the documents show disclosed seats and published funding lines, not coordination.

Alignment Analysis

Stated alignment: AIUC’s mission, on its team page: “Our mission is to Underwrite Superintelligence. We believe progress requires security, incentives must be shaped, AI risk is a new version of an old problem, and markets can help govern AI” (AIUC, team). His own, on AVERI’s launch day: “We need 3rd party audits & verification to unlock the insurance market and create confidence in AI adoption” (Dattani, X). And on the proper order of private and public bodies: “the ecosystem will rise much faster than government can get started. Private verifiers & labs can experiment now to build the right oversight techniques” (Dattani, X). Observed alignment: Consistent. The company, the essays, the posts and the board seat all point at the same mechanism: audits that an insurer reads and prices. Gap assessment: None documented in the words. The structural question is the one on the page: a METR board member’s company proposes a lab mutual that would hire METR-type evaluators (“qualified third parties such as Accenture or METR”), while METR’s own policy states “To date, we have not received payment for work on company-identifying risk assessments” and its president wrote on 25 September 2026 that “it’s a bad idea to have frontier AI evaluators get paid by AI companies directly today.” No page says METR has been engaged or paid by AIUC, an insurer or a mutual. The seat is disclosed on both organizations’ pages (Lawfare; METR, conflict of interest policy v1.1; Painter, X).

Interest: Four positions, each in his own words.

• Insurers choose and pay evaluators. Position, 25 September 2026, replying to METR’s president: “Insure the labs and have the insurer choose & pay evaluators. Insurer’s incentives are to reduce claims, so demand rigorous evaluators” (Dattani, X); in Lawfare, audits “selected, contracted, and paid for by AIMIL” (Lawfare). What adoption would pay him: the company he co-founded lists audits, standards and insurance as its lines of work and proposes the body that would buy the audits; no page says AIUC would supply, staff or run the mutual, and his equity is not published. The METR board he sits on belongs to the type of evaluator named as an example contractor; no payment to METR is documented. Best case, his: insurers “demand rigorous evaluators” because they pay the claims: “If you’re paying when a boiler explodes, it helps to know why boilers explode” (Dattani, X). Who else gains: the evaluators and auditors a mutual would hire, and the labs, whose liability becomes priced and pooled.
• Insurance priced on audit results. Position: “Nuclear Electric Insurance Limited uses independent plant evaluations to set premiums: better evaluations earn discounts” (Dattani, X); on ElevenLabs, “Certifying against AIUC-1 gave insurers the confidence they needed” (Dattani, X). What adoption would pay him: each buyer that asks vendors for an AIUC-1 certificate buys the audit AIUC conducts and issues; AIUC-1’s site tells buyers “Ask your AI vendors if they are AIUC-1 certified” (AIUC-1). Fees are not published. Best case, his: “Insurance without that testing doesn’t solve the problem” (Dattani, X). Who else gains: Schellman, KPMG and the other auditors; the vendors that want a trustmark.
• Third-party selection of auditors, not mandatory audits. Position: “I don’t think mandatory audits solves this problem – we need a 3rd party to select auditors” (Dattani, X). What adoption would pay him: a third party that selects and contracts auditors is the role an insurer or mutual fills; a mandate without that selector leaves the labs choosing. Best case, his: a mutual has “legible incentives to pick high quality evaluators” (Dattani, X). Who else gains: AVERI, which argues for mandatory audits, is a different design and lists AIUC among its funders; the two converge on third-party auditing and differ on who selects.
• Private verifiers first; a mandate for insurance later. Position: “If the regime proves itself, it could be strengthened and spread by mandating insurance” (Lawfare, co-written). What adoption would pay him: a mandated insurance requirement makes a voluntary certificate a condition of operating. Best case, his: “the ecosystem will rise much faster than government can get started.” Convergence: Fathom’s California designation law and AVERI’s audit advocacy ask government for a related, differently built mandate.

Interest is a documented fact about each position, not a claim about motive, and not a verdict that the argument is wrong. Each best case is stated at the strength its author states it.

Convergent Drive Classification

Self-preservation: Holds a founder’s seat at a venture-funded company with $55 million raised, an advisory and board seat at METR, and (until FY2024) a paid COO and Treasurer seat there (AIUC, Series A; ProPublica). Goal preservation: Keeps the argument for insurance-priced audits in front of three audiences at once: enterprise buyers through AIUC-1, evaluators through the METR seat and the Forum letter, and policy readers through Lawfare. Resource acquisition: AIUC’s $15 million seed and $40 million Series A, with the frontier-model market as the stated next use of the proceeds (AIUC, Series A). Self-improvement: From insurance partner to evaluator operations chief to founder of a company that sells the standard and the insurance channel.

Subject is not an AI system. The drives appear anyway — in the underwriter’s seat.

Sources: AIUC, team; AIUC, Series A; AIUC, KPMG; AIUC, Schellman; AIUC, AI-Proofing The Board and C-suite; AIUC-1; AIUC-1, consortium; AIUC-1, CSA STAR; Lawfare, Bootstrapping Frontier AI Governance by Mutualizing Risk; METR, About; METR, Rajiv Dattani; METR, conflict of interest policy v1.1; ProPublica, Model Evaluation And Threat Research; AVERI, Frontier AI Auditing; AVERI, About; Halcyon Futures; AI Evaluator Forum, letter; GovAI, Underwriting the Agent Economy; SiliconANGLE, 15 September 2026; Dattani on X, @RajivDattani; Painter on X.