Skip to main content
v2026.11,858 entries · CC-BY 4.0
NIKOLAI elementN8 · Transparency and reviewProposednikolai-v0.1

Evaluator Independence and Conflict of Interest

NIKOLAI proposal: Evaluator Independence and Conflict of Interest is a record of declared financial, organisational and personal relationships between an evaluator and the developer being evaluated, and the independence test applied to clear the evaluator for the engagement. This working definition is NIKOLAI's own editorial synthesis, corresponding to the ALIGNMENT-MATRIX.md source document's added element "H5 Evaluator independence and conflict of interest."

This is CASRAI's own proposed definition, not a definition any named organisation has agreed to. See what NIKOLAI is and is not.

Source of record

Where this definition comes from

  • Anthropic Responsible Scaling Policy v3.4, §3.6.1, Selecting external reviewers

    significant expertise in evaluations for dangerous AI capabilities and propensities, including awareness of ways evaluations 'might be misleading (for example, alignment faking)'; ... a reviewing organization itself may not have a financial interest in Anthropic; and the individuals involved in conducting the review, as well as anyone above them in the reporting chain within their organization, may not have a financial interest in Anthropic or close personal relationships with anyone at Anthropic (i.e., family relationships, romantic relationships, or shared living arrangements).

    https://cdn.sanity.io/files/4zrzovbb/website/0bacdc8440ea96e62a8766d99ebe1d4eea6d5f3a.pdf
  • METR: OpenAI-Hugging Face incident investigation, conflict-of-interest footnotes

    METR does not accept compensation for this work. Ryan Greenblatt is the domestic partner of Beth Barnes, METR's CEO. Beth Barnes was not involved in the decision to engage Ryan Greenblatt, nor was she directly involved in this investigation.

    https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/
  • FRONTIER Act, 119th Congress (H.R. 9925, introduced, not enacted), Sec. 4(c); Sec. 3(c)

    independent, with no financial interest either way" (Sec. 4(c)); IVOs must "show independence and meet funding-transparency and conflict-of-interest rules" as a licensing condition (Sec. 3(c)).

    https://www.govinfo.gov/content/pkg/BILLS-119hr9925ih/html/BILLS-119hr9925ih.htm

Crosswalk

How named organisations use this concept

Every row below is a shadow mapping. It is CASRAI's own reading of a published document. No lab, evaluator or regulator named here has declared, endorsed, or been consulted on this mapping. That will change only when an organisation files its own Mapping Declaration — see the non-endorsement policy.
OrganisationTheir term, as publishedMatchSource
Anthropic
Anthropic Responsible Scaling Policy v3.4 / Advanced AI Framework / Amodei essay
RSP v3.4 §3.6.1, three tests: significant expertise in "evaluations for dangerous AI capabilities and propensities", including awareness of ways evaluations "might be misleading (for example, alignment faking)"; incentives to be candid, e.g. "external review parties should not be teams whose revenue, reputation and success depend entirely on Anthropic and similar companies continuing to work with them"; and no conflicts of interest -- "a reviewing organization itself may not have a financial interest in Anthropic; and the individuals involved in conducting the review, as well as anyone above them in the reporting chain within their organization, may not have a financial interest in Anthropic or close personal relationships with anyone at Anthropic." Selection requires Board consultation and LTBT approval (§3.6.1). AAF: "independent in the sense of not having a financial interest in the developer and being free of major conflicts of interest with respect to the individuals involved in conducting the review" (p.7); "Evaluator shopping." Amodei: "a second opinion free of commercial incentives."
Also cites {AAF}, {PACE}.
exact
confidence: high
Anthropic Responsible Scaling Policy v3.4
EU
EU GPAI Code of Practice, Safety and Security Chapter
Appendix 3.5 "adequate qualification" test: "(1) having significant domain expertise for the systemic risk and being technically skilled and experienced in conducting model evaluations; (2) having appropriate internal and external information security protocols in place; and (3) having agreed to protect commercially confidential information, if they need access to such information." No explicit financial-conflict-of-interest test (unlike Anthropic's RSP §3.6.1); Model Report must give "an explanation of the choice of evaluator based on the qualification criteria" (Measure 7.4).
Notably lacks an explicit financial-COI test, unlike Anthropic's RSP.
close
confidence: medium
EU GPAI Code of Practice, Safety and Security Chapter
METR
METR (site) / METR OpenAI-Hugging Face incident investigation
"METR does not accept compensation for this work." Per-person COI footnotes: "Ryan Greenblatt is the domestic partner of Beth Barnes, METR's CEO. Beth Barnes was not involved in the decision to engage Ryan Greenblatt, nor was she directly involved in this investigation." METR accepted "~$400K in free API credits."
Also cites {METRHF}.
exact
confidence: high
METR
Frontier Model Forum
FMF Third-Party Assessments technical report
"Methodological independence: ... enables assessors to approach evaluations with a fresh perspective"; open-ended assessment may involve "internal assessors who are sufficiently independent"; brief gap: "No criteria for independence (financial, contractual, organisational)."close
confidence: medium
FMF Third-Party Assessments
AI Evaluator Forum (AEF-1, discovery sweep)
Discovery sweep
AEF-1 condition "Minimized Conflicts of Interest" (discovery).
Unverified discovery-stage material.
close
confidence: low
AI Evaluator Forum (AEF-1) discovery sweep
US Congress (H.R. 9925, FRONTIER Act, not enacted)
FRONTIER Act, 119th Congress
Compliance auditors must be "independent, with no financial interest either way" (Sec. 4(c)); IVOs must "show independence and meet funding-transparency and conflict-of-interest rules" as a licensing condition (Sec. 3(c)), and the licensing regulations must cover "conflict-of-interest and funding-transparency requirements, including reporting requirements regarding the IVOs' funding sources and revenue generation." A named "lead assessment partner" certifies "freedom from conflicts" on each report (Sec. 5).
Not enacted.
exact
confidence: high
FRONTIER Act, 119th Congress (H.R. 9925, introduced, not enacted)
US Congress (S. 5061, 119th Congress, not enacted)
S. 5061, 119th Congress
The proposed AI Risk Board's conflict-of-interest policy: members have "a fiduciary responsibility to the Board, a duty to report conflicts of interest, including the appearance of a conflict of interest, and do not participate in deliberations or votes from which they personally or their employer would directly and materially benefit" (Sec. 3(b)(5)(D)(i)); members "who breach the COI policy are removed" (Sec. 3(b)(3)); lab-affiliated experts may sit on the Board "subject to COI disclosure and recusal"; members are sponsored for TS/SCI clearance, and "members denied a clearance, or restricted from classified information, are removed" (Sec. 3(b)(4)).
An institutional (board-member) COI regime rather than a reviewer-selection COI regime; described in the source document as a fuller independence regime than Anthropic's RSP §3.6.1 on institutional grounds. Not enacted.
close
confidence: medium
S. 5061, 119th Congress

Related, not mapped

Pointers that are not crosswalk claims

These sources mention this concept but do not define or map it clearly enough to count as a crosswalk row — noted here so the research is visible without overstating it as a mapping.

  • OpenAI

    Proposal that CAISI certify third-party assessors: "independent assessment certification" (blueprint p.5) -- RL, a pointer not a mapping; this is an OpenAI proposal, not current CAISI practice.

    A Blueprint for a Federal Framework
  • Google DeepMind

    HuBREC: "a research ethics committee run within Google DeepMind but staffed and chaired by academics from outside the company" (p.21) -- RL, a pointer not a mapping.

    Gemini 3.7 Flash FSF report
  • Meta

    Evaluations "held out from teams conducting training to reduce the risk of overfitting" (§4.1) -- internal segregation, RL, a pointer not a mapping.

    Meta Advanced AI Scaling Framework v2
  • US Government

    "See OAI (CAISI certification is an OpenAI proposal, not current CAISI practice)" -- RL, a pointer not a mapping.

    A Blueprint for a Federal Framework

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →