Skip to content
GLOBALANCE INTELLIGENCEThe economy ahead.
This Week Index 28/100 Sign in
Money & Investments › GI 100 Index
The Great Value ShiftInvestment ThemesGI 100 IndexGI PortfolioStart Future Check
AI FOOTPRINTlimitedASSESSMENT 2026GLOBALANCEINTELLIGENCE
Globalance AI Footprint

Assessment as of November 2026 · next assessment 2027.

Result of the assessment by the Globalance research team under the published methodology. Not a certification or quality label.

How we assess ›

How this rating is builtVideo · 0:52
GI 100 Index · Profile

Bristol-Myers Squibb Company

USAAdoptersAssignment FullData quality Medium

Key messagelimited AI Footprint

Bristol Myers Squibb has anchored oversight of artificial intelligence with its board and two of its committees and describes a regularly externally assessed programme to protect its IT systems. However, the company has not published binding rules of its own for dealing with artificial intelligence, and it discloses very little about how individual AI applications, for example in drug safety or manufacturing, are tested and where their limits lie.

Disclosure: the research was supported by an AI model made by Anthropic. Bristol-Myers Squibb Company is connected to Anthropic (e.g. as investor, partner, supplier, customer or through people). The assessment was therefore reviewed by people.

Opportunities

  • Safetysolid

    Item 1C of the 10-K describes a concrete cybersecurity programme with an independent assessment against NIST every three years, an incident response plan with several tabletop exercises per year, supplier risk management and oversight by the Audit Committee; material incidents in three years are …

    Item 1C of the 10-K describes a concrete cybersecurity programme with an independent assessment against NIST every three years, an incident response plan with several tabletop exercises per year, supplier risk management and oversight by the Audit Committee; material incidents in three years are denied (S02, S03, grade B). For the AI itself, reliability and testing are documented only in places: the WEF confirms the productive use of AI in Devens (S17, grade A, without reference to safety), and the FDA clearance of the radiology algorithms is reported only by secondary sources (S18). Misuse and incident processes specific to AI (bias, drift) are mentioned only as tasks in job postings (S13), and the validation of AI in pharmacovigilance remains open (S05).

Risks

  • Responsibilitylimited

    The 2026 proxy statement explicitly places oversight of the use of AI with the board and assigns it to the Science and Technology Committee for R&D and to the Audit Committee for cybersecurity and data protection (S03, grade B); at management level, the Chief Digital & Technology Officer, the CISO …

    The 2026 proxy statement explicitly places oversight of the use of AI with the board and assigns it to the Science and Technology Committee for R&D and to the Audit Committee for cybersecurity and data protection (S03, grade B); at management level, the Chief Digital & Technology Officer, the CISO and the CPO are responsible (S02, S03). However, an AI policy with operational content published by the company, or a described approval, risk classification or review procedure for AI, is missing: the code of conduct sets out principles only (S06, grade C), and the AI governance programme (model inventory, risk review, human oversight) is documented only through job postings (S12, S13), which do not count for this; neither an AI policy nor an AI position paper is available on the website (S27, S28). Because board oversight alone is not sufficient and figures, certifications or external audits of the AI programme are missing, the pillar is “limited” (principles without evidence of implementation).

  • Transparencylimited

    BMS describes many AI applications publicly (S05, S14, S24) and discloses in its privacy notice the use of AI in applications and automated decisions (S07).

    BMS describes many AI applications publicly (S05, S14, S24) and discloses in its privacy notice the use of AI in applications and automated decisions (S07). For none of its own systems, however, are performance limits, error rates, validation results or model information published; the Impact Report promises transparency only as a principle (S05, S06). The information therefore rests on grade C evidence.

  • Risk area

    AI in drug safety and in manufacturing

  • Risk area

    Patient-facing radiology algorithms with wide reach in US hospitals

  • No serious controversy recorded.

What lies behind the assessment

Research of 3 November 2026
42searches
31pages and documents read
11primary documents
1controversies checked
6open points, named openly
The full research: pillars, evidence, controversies and open points
AI Footprint · Adopter · As of 3 November 2026
Bristol-Myers Squibb CompanyOverall rating: limited

Bristol Myers Squibb has anchored oversight of AI with its board and has its cybersecurity regularly assessed externally. It has not published its own AI rules and discloses little about the testing and limits of its AI.

Responsibility

40 %
limitedsolidstrong

The board oversees AI, but published AI rules and review procedures are missing.

  • +Proxy statement places AI oversight with the board Source ↗
  • +2026 code of conduct newly contains a section on AI Source ↗
  • –AI governance programme documented only through job postings Source ↗
  • –No published AI policy and no described review procedure Source ↗
  • –No certification or external audit of the AI programme
All sources for this pillar (12)

Safety

40 %
limitedsolidstrong

The cybersecurity programme is documented concretely, but testing of the AI itself only in places.

  • +Independent assessment of cybersecurity against NIST every three years Source ↗
  • +Technicians decide on AI recommendations in running batches Source ↗
  • –Validation of AI in pharmacovigilance not documented Source ↗
  • –FDA clearance of the radiology algorithms reported only by secondary sources Source ↗
  • –Data leak at distributor Cencora involving health data Source ↗
All sources for this pillar (10)
What is missing for “strong”

These steps are in the company’s own hands.

1Responsibility
Publish the AI policy and have it audited

Publish an AI policy with responsibilities, risk classes and approval procedures, document its implementation with figures and have it externally audited.

2Safety
Document validation of AI in GxP processes

Disclose validation and monitoring of AI in pharmacovigilance and manufacturing.

3Transparency
Disclose the limitations of its own models

Publish system cards with limitations, error rates and training data for the radiology algorithms and Predict First models.

Context: 101 companies assessed

Distribution of overall ratings, no ranking
strong
none yet
solid
limited
← Bristol-Myers Squibb Company is one of 66

No company reaches “strong” today. Disclosure on how companies handle AI is still young across the market.

Controversies
1 1 minor

No deduction for controversies.

Largest risk area
AI errors in GxP processes

No validation or monitoring procedures are documented for generative AI in pharmacovigilance or for AI recommendations in manufacturing.

How thoroughly assessed
31 Evidence

11 primary documents read, 42 searches.

Read the full reasoning and all sources
Bristol-Myers Squibb Company · Full reasoning

Board oversight of AI, but no published AI rules and little information on testing

Overall rating “limited”. No deduction for controversies. The superscript numbers link directly to the source.

1 · Responsibility

limitedWeight 40%

Oversight of AI is documented in the proxy statement, but a published AI policy or a described review procedure is missing; the programme is documented only through job postings.

What is documented
  • The board's risk oversight covers the use of AI; the Science and Technology Committee oversees AI in research and development2
  • The Audit Committee oversees information security and data protection and receives reports from the CISO2 1
  • The CISO reports to the Chief Digital & Technology Officer1
  • The 2026 code of conduct commits to responsible and ethical use of AI; the 2018 edition did not yet contain this4 8
  • The 10-K mentions investment in AI upskilling and training on the responsible use of AI tools9
The case against
Programme documented only indirectly
  • Model inventory, risk classification and life-cycle controls are described only in job postings; their implementation status is open5 3
  • Neither an AI policy nor a position paper on AI is available on the BMS website6 7
Missing accountability
  • Training rates, AI figures, a certification or an external audit of the AI programme are missing
  • The Integrity Line is a general reporting channel; there is no AI-specific complaints channel with documented remedy4
Sources for Responsibility (12)
  1. Board Cybersecurity (board-cybersecurity.com): BRISTOL MYERS SQUIBB CO 10-K Item 1C. Cybersecurity – 2026-02-11 · 11 February 2026 · primary source
  2. Bristol-Myers Squibb Company (SEC EDGAR): Definitive Proxy Statement (DEF 14A) 2026, Annual Meeting 2026-05-05 · 25 March 2026 · primary source
  3. Mediabistro (Job Posting Bristol Myers Squibb): Senior Director, AI Governance, Law & Compliance (R1598558) · 3 November 2026 · secondary source
  4. Bristol-Myers Squibb Company: 2026 Principles of Integrity: BMS Standards of Business Conduct and Ethics · 1 January 2026 · secondary source
  5. Legal.io (Job Posting Bristol Myers Squibb): Counsel, AI Governance, Law and Compliance · 3 November 2026 · secondary source
  6. Bristol-Myers Squibb Company: Transparency & reporting · 3 November 2026 · secondary source
  7. Bristol-Myers Squibb Company: Position on key issues · 3 November 2026 · secondary source
  8. Bristol-Myers Squibb Company (über EthicsPoint): Principles of Integrity: Standards of Business Conduct and Ethics (edition 2/2018) · 1 February 2018 · secondary source
  9. Bristol-Myers Squibb Company (SEC EDGAR): Form 10-K for the fiscal year ended December 31, 2025 · 11 February 2026 · primary source
  10. Bristol-Myers Squibb Company: 2025 Impact Report · 31 December 2025 · secondary source
  11. Bristol-Myers Squibb Company: Privacy Notice (bms.com) · 3 November 2026 · secondary source
  12. Bristol-Myers Squibb Company (Kopie über Fortune/Quartr): Form 10-K FY2025 (PDF version) · 11 February 2026 · primary source

2 · Safety

solidWeight 40%

BMS documents a cybersecurity programme assessed externally against NIST, but the validation of AI in pharmacovigilance remains open.

What is documented
  • Item 1C describes an independent assessment against NIST every three years, an incident response plan with exercises and supplier risk management6
  • The CPO and CISO run a joint data protection programme; the board is informed at least once a year6 7
  • The WEF recognised the Devens site with more than 30 digital and AI applications as a Lighthouse4
  • In Devens and in “Predict First”, specialists make the final decision on AI recommendations3 2
The case against
Missing evidence of AI testing
  • There is no information on validation, red-teaming and monitoring of AI in pharmacovigilance1
  • Incident processes for AI bias and drift are mentioned only as a task in a job posting9
Incident at a service provider
  • In 2024, health data of customers of BMS programmes was stolen at distributor Cencora; according to the report, BMS systems were not affected10
  • Protective measures for models, training data and access by AI agents are not disclosed
Sources for Safety (10)
  1. Bristol-Myers Squibb Company: 2025 Impact Report · 31 December 2025 · secondary source
  2. PharmaVoice (Alexandra Pecci): How Bristol Myers Squibb uses AI to validate drugs before the lab («Predict First») · 24 July 2025 · secondary source
  3. The New York Times (Farah Stockman), reprinted in The Star: American factories lag in adopting AI. This drugmaker is an exception. · 9 May 2026 · secondary source
  4. World Economic Forum: Global Lighthouse Network Recognizes 23 New Sites, Launches AI Platform for Industrial Transformation · 15 January 2026 · primary source
  5. AuntMinnie: Bristol Myers Squibb, Microsoft partner on early lung cancer detection · 20 January 2026 · secondary source
  6. Board Cybersecurity (board-cybersecurity.com): BRISTOL MYERS SQUIBB CO 10-K Item 1C. Cybersecurity – 2026-02-11 · 11 February 2026 · primary source
  7. Bristol-Myers Squibb Company (SEC EDGAR): Definitive Proxy Statement (DEF 14A) 2026, Annual Meeting 2026-05-05 · 25 March 2026 · primary source
  8. Bristol-Myers Squibb Company: 2026 Principles of Integrity: BMS Standards of Business Conduct and Ethics · 1 January 2026 · secondary source
  9. Mediabistro (Job Posting Bristol Myers Squibb): Senior Director, AI Governance, Law & Compliance (R1598558) · 3 November 2026 · secondary source
  10. TEISS: Pharmaceuticals firm Cencora says data breach impacted Bristol Myers Squibb customers · 23 May 2024 · secondary source

3 · Transparency

limitedWeight 20%

BMS describes its AI applications publicly but publishes no performance limits, error rates, validation results or model information for any of its own systems.

What is documented
  • The Impact Report and specialist articles describe AI in research, clinical development, manufacturing and drug safety2 4 5
  • The privacy notice mentions AI-supported candidate selection with a final human decision and possible automated decisions3
  • Transparency and ethical oversight of AI are named as principles in the Impact Report and the code of conduct2 6
The case against
Missing system information
  • Results of “Predict First”, figures from Devens and models for protein degradation are described without methodology or error rates4 7 5
  • There are no model or system cards and no information on training data for the radiology algorithms and Predict First models
Missing information for those affected
  • Whether patients and physicians are informed about the use of AI is not documented
  • No register of the AI systems in use was found
Sources for Transparency (9)
  1. Bristol-Myers Squibb Company (SEC EDGAR): Form 10-K for the fiscal year ended December 31, 2025 · 11 February 2026 · primary source
  2. Bristol-Myers Squibb Company: 2025 Impact Report · 31 December 2025 · secondary source
  3. Bristol-Myers Squibb Company: Privacy Notice (bms.com) · 3 November 2026 · secondary source
  4. PharmaVoice (Alexandra Pecci): How Bristol Myers Squibb uses AI to validate drugs before the lab («Predict First») · 24 July 2025 · secondary source
  5. Bristol-Myers Squibb Company: Accelerating protein degradation science with AI and machine learning · 31 July 2024 · secondary source
  6. Bristol-Myers Squibb Company: 2026 Principles of Integrity: BMS Standards of Business Conduct and Ethics · 1 January 2026 · secondary source
  7. The New York Times (Farah Stockman), reprinted in The Star: American factories lag in adopting AI. This drugmaker is an exception. · 9 May 2026 · secondary source
  8. Intelligent CIO: Bristol Myers Squibb builds AI factory with NVIDIA to accelerate drug discovery · 31 July 2026 · secondary source
  9. Bristol-Myers Squibb Company: Bioethics policy statement · 3 November 2026 · secondary source

4 · Why the harm rule does not apply

The harm rule does not apply: in the research period no judgment, no authority decision, no fine and no pattern of harm from BMS's use of AI was found.

5 · Controversies (1)

6 · What remains open

  • Public AI policy, risk classes and approval procedures for in-house and purchased AI
  • Implementation status of the AI governance framework with model inventory, reviews and figures
  • Validation and monitoring of AI in pharmacovigilance and manufacturing
  • AI-related risk factors and legal proceedings from Items 1A and 3 of the 10-K, which were not readable
  • FDA primary evidence of the clearance of the radiology algorithms
  • Model or system cards and information on the limitations of its own models

7 · How it was assessed

Cut-off date: 3 November 2026 · 42 searches, 31 fetches.

Primary documents read (11)
  • Form 10-K FY2025 (Item 1 through start of Item 1A, Human Capital)
  • Item 1C Cybersecurity 10-K FY2025 (via reproduction S02)
  • Proxy Statement DEF 14A 2026 (through Director Compensation, two retrievals)
  • DEFA14A Proxy Season Update 2025
  • Form 8-K Voting Results 2026
  • Form 10-Q Q1 and Q3 2026 (Notes, partial)
  • 2025 Annual Report (ARS), Shareholder Letter
  • 2025 Impact Report (pp. 1–44)
  • 2026 Principles of Integrity and 2018 Edition
  • Privacy Notice bms.com (partial)
  • Bioethics policy statement

Research of 3 November 2026 under Globalance AI Footprint rulebook version 0

We assess documented practice in dealing with AI; this is not an investment recommendation. Three pillars, weights 40/40/20, harm rule.

Research of 3 November 2026 under Globalance AI Footprint rulebook version 0. Documented observations from public sources, no investment recommendation.

Important Notice. This material is provided for information purposes only and does not constitute an offer or solicitation to buy or sell any financial instrument, nor does it constitute investment advice or a recommendation. The GI 100 Index is a rules-based strategy index of Globalance Bank. Investments involve risk, including the loss of capital invested. The final product documents (including FinSA) shall prevail in all cases.

Your company? Information on inclusion in the GI 100 Index →

What comes next