AI committee on the board, but only principles on AI review and health data
Overall rating “limited”. No deduction for controversies. The superscript numbers link directly to the source.
1 · Responsibility
Board oversight of AI is documented in SEC filings, but review and approval of AI rest only on a position statement, and according to a job posting the risk taxonomy is still being built.
What is documented
- Since April 2026 the Digital Technology Committee has been responsible for the strategic use of AI and the governance framework and meets at least twice a year1 2 3
- Previously the Audit Committee named AI and digital technology as a focus; one director brings AI expertise2 1
- The position statement requires review of new AI solutions before approval, human review of outputs and supplier due diligence4
- At management level there is an AI Advisory Committee and the position of a Chief AI Officer4 5
- General reporting channels with a ban on retaliation exist; 99.6% completed code of conduct training in 20257 8
The case against
Controls still being built
- A job posting from October 2026 only now plans to establish a risk taxonomy, classification standards and a lifecycle standard5
- An AI policy with risk classes and a described approval process are not published
Missing accountability
- No AI report, AI metrics, AI training rate or external audit was found; the 2024 and 2025 sustainability reports could not be retrieved
- The grievance channels are general, without reference to AI and without documented remedy7
Sources for Responsibility (9)
- Regeneron Pharmaceuticals, Inc. (SEC EDGAR): Definitive Proxy Statement (DEF 14A) 2026, Annual Meeting 12 June 2026 · 24 April 2026 · primary source
- Regeneron Pharmaceuticals, Inc. (SEC EDGAR): Proxy Statement 2026, courtesy PDF · 24 April 2026 · primary source
- Regeneron Pharmaceuticals, Inc. (Investor Relations): Charter of the Digital Technology Committee of the Board of Directors · 3 November 2026 · primary source
- Regeneron Pharmaceuticals, Inc.: Position Statement on Responsible Use of Artificial Intelligence · 1 July 2025 · secondary source
- Regeneron Pharmaceuticals, Inc. (careers page): Executive Director, Responsible AI & Enterprise Governance (R49699) · 14 October 2026 · secondary source
- Regeneron Pharmaceuticals, Inc. (SEC EDGAR, XBRL): 10-K FY2025, R29: Cybersecurity Risk Management and Strategy Disclosure (Item 1C) · 4 February 2026 · primary source
- Regeneron Pharmaceuticals, Inc.: Code of Business Conduct and Ethics · 7 October 2026 · secondary source
- Regeneron Pharmaceuticals, Inc.: Achieving Our 2025 Responsibility Goals · 31 December 2025 · secondary source
- Regeneron Pharmaceuticals, Inc.: Regeneron Named to Dow Jones Sustainability World Index for Sixth Consecutive Year · 16 December 2024 · secondary source
2 · Safety
Regeneron describes in the 10-K a cybersecurity programme with a CISO, external assessors and committee oversight, but for validation and monitoring of AI there are only company statements.
What is documented
- Item 1C describes risk assessment, monitoring and testing of systems, external assessors, supplier due diligence and reports to the Audit Committee4
- A Technology Risk Management Committee assesses technology risks within the enterprise risk programme4
- Since April 2026 the Digital Technology Committee has overseen the data breach response plan and the data governance framework5 6
- The Regeneron Genetics Center prohibits re-identification attempts; the TriNetX matching uses «privacy-preserving methods»7 8
- The position statement requires protection against data leakage when using AI with confidential information1
The case against
Missing evidence of AI testing
Sources for Safety (11)
- Regeneron Pharmaceuticals, Inc.: Position Statement on Responsible Use of Artificial Intelligence · 1 July 2025 · secondary source
- Regeneron Pharmaceuticals, Inc.: From Human Data to Human Health (Regeneron Genetics Center) · 3 November 2026 · secondary source
- Regeneron Pharmaceuticals, Inc. (careers page): Executive Director, Responsible AI & Enterprise Governance (R49699) · 14 October 2026 · secondary source
- Regeneron Pharmaceuticals, Inc. (SEC EDGAR, XBRL): 10-K FY2025, R29: Cybersecurity Risk Management and Strategy Disclosure (Item 1C) · 4 February 2026 · primary source
- Regeneron Pharmaceuticals, Inc. (SEC EDGAR): Proxy Statement 2026, courtesy PDF · 24 April 2026 · primary source
- Regeneron Pharmaceuticals, Inc. (Investor Relations): Charter of the Digital Technology Committee of the Board of Directors · 3 November 2026 · primary source
- Regeneron Pharmaceuticals, Inc.: Regeneron Genetics Center Data Privacy Statement · 1 May 2025 · secondary source
- Regeneron Pharmaceuticals, Inc. (GlobeNewswire): Regeneron Collaborates with TriNetX to Access De-identified Electronic Health Records of 300 Million Patients · 2 April 2026 · secondary source
- Board Cybersecurity: Regeneron Pharmaceuticals, Inc. Cybersecurity Incident (Tracker) · 22 May 2024 · secondary source
- Claim Depot: Regeneron Pharmaceuticals, Inc. Data Breach · 24 May 2024 · secondary source
- BleepingComputer: Cencora data breach exposes US patient info from 11 drug companies · 24 May 2024 · secondary source
3 · Transparency
Regeneron names the types of AI used and its principles, but discloses neither limits nor consent information on its own models and the use of health data for AI training.
What is documented
- The position statement names machine learning, language processing, large language models and process automation as types of AI used1
- Regeneron commits to labelling AI interaction «where possible», disclosing AI-generated content in publications and documenting decisions of AI solutions1
- The position statement prohibits letting AI replace critical decisions such as hiring1
- Regeneron names its data holdings: over 3 million exomes, data on around 300 million people via TriNetX and 10 million exomes at Truveta4 2 5
The case against
Sources for Transparency (5)
- Regeneron Pharmaceuticals, Inc.: Position Statement on Responsible Use of Artificial Intelligence · 1 July 2025 · secondary source
- Regeneron Pharmaceuticals, Inc. (GlobeNewswire): Regeneron Collaborates with TriNetX to Access De-identified Electronic Health Records of 300 Million Patients · 2 April 2026 · secondary source
- Regeneron Pharmaceuticals, Inc.: Regeneron Genetics Center Data Privacy Statement · 1 May 2025 · secondary source
- Regeneron Pharmaceuticals, Inc.: From Human Data to Human Health (Regeneron Genetics Center) · 3 November 2026 · secondary source
- Fortune: Truveta Genome Project with Regeneron and Illumina · 16 January 2025 · secondary source
4 · Why the harm rule does not apply
The harm rule does not apply: within the window there is no judgment, authority decision or fine on harm from Regeneron's AI; the service provider incident is unrelated to AI.
5 · Controversies (1)
6 · What remains open
- AI-related risk factors in the FY2025 10-K, whose Item 1A was truncated on retrieval
- Design and status of the AI risk framework with risk taxonomy, approval process and AI inventory
- Validation and monitoring of AI in GxP processes
- Consent and governance in the use of TriNetX and Truveta data for AI training
- Content of the 2024 and 2025 sustainability reports on AI, data protection and cybersecurity
7 · How it was assessed
Cut-off date: 3 November 2026 · 42 searches, 32 fetches.
Primary documents read (10)
- Form 10-K FY2025 (Item 1 through Government Regulation; Item 1C via R29; Note 16 Legal Matters via R23)
- Proxy Statement 2026 (HTML and courtesy PDF, each up to approximately 117,000 and 100,000 characters respectively)
- Charter of the Digital Technology Committee
- Position Statement on Responsible Use of Artificial Intelligence (July 2025)
- Code of Business Conduct and Ethics (October 2026)
- Achieving Our 2025 Responsibility Goals
- RGC Data Privacy Statement (May 2025)
- Form 8-K Voting Results 2024, 2025, 2026
- Form 10-Q Q2 2026 (to Note 6)
- Press releases: TriNetX (2026-04-02) and 23andMe (2025-05-19)
Research of 3 November 2026 under Globalance AI Footprint rulebook version 0