Bristol Myers Squibb AI Strategy: Predict First, Then Go to the Lab
How Bristol Myers Squibb uses AI on its own work: Claude for 30,000+ employees, AI predictions before bench work and a goal to halve discovery time. Updated November 2026.
Prescription medicines in oncology, hematology, immunology, cardiovascular disease and neuroscience
BMS has no standalone AI strategy document. Its approach comes from its May and July 2026 announcements and CEO remarks on an April 2026 investor call.
Read the company sourceWhere Bristol Myers Squibb Is Going with AI
Bristol Myers Squibb wants AI to change how the whole company operates. In May 2026, it made Claude, from Anthropic, a shared AI platform for more than 30,000 employees across research, clinical development, manufacturing, commercial and corporate functions. BMS has no standalone AI strategy document, so this profile draws on company releases and executive remarks.
The goal goes beyond chat. Greg Meyers, Executive Vice President and Chief Digital and Technology Officer, said: “Most enterprise AI stops at the chatbot.” He told FiercePharma: “The real prize is the untapped value still trapped behind decades of data silos.”
Research runs on a principle called Predict First. AI predictions come before bench work, and BMS says the approach now informs every small-molecule program and most large-molecule programs (PharmExec, July 2026). CEO Christopher Boerner has set a goal to halve the time from target selection to lead molecule, according to FiercePharma.
The focus has five parts.
- Research: AI co-scientists and foundation models trained on BMS data
- Clinical development: shorter time from data lock to regulatory filing
- Manufacturing and quality: Claude tested as an agent layer in workflows
- Commercial and medical affairs: AI agents in field and medical workflows
- Software: Claude Code for engineering and data science teams
What It Takes: One Shared Platform, Several Vendors and Its Own Compute
BMS has been building for more than three years. It launched its first internal AI chatbot in January 2023 and gives employees unlimited access to frontier models through a proprietary platform. It runs a multi-vendor strategy, with Claude as the shared layer.
BMS is also building its own compute. In July 2026, it announced an NVIDIA DGX SuperPOD with Vera Rubin systems to train foundation models on its proprietary data. Meyers said: “BMS has made a deliberate bet on AI, and we are beginning to see it pay off in our pipeline and operations.”
Humans stay in charge of decisions. BMS calls its model hybrid intelligence, in which AI co-scientists do data-heavy tasks while scientists direct interpretation and decisions. Robert Plenge, Executive Vice President and Chief Research Officer, said the aim is “raising the probability that each program we advance is the right one.”
Governance is built into the platform. BMS says its Claude integrations run with full enterprise governance and audit controls. It has published no broader AI governance framework.
BMS is cutting costs at the same time. A strategic productivity initiative announced in early 2025 targets $2 billion in savings by the end of 2027, on top of an earlier $1.5 billion program (FiercePharma, 2025). BMS has not tied the cuts to AI, and headcount fell by about 1,600 in 2025.
AI fluency, coded against the Lead with AI Fluency Matrix (September 2026 edition):
- Fluency floor: Not published. All employees have access to frontier models and more than 30,000 are getting Claude, but BMS has set no required level for all staff.
- Highest level evidenced: Level 3 (Integrating), company-claimed. Predict First builds AI predictions into every small-molecule program, and BMS says AI agents save scientists weeks of manual work in target identification. These claims have not been independently verified.
- Portfolio design: BMS puts one shared AI layer across all functions and invests most heavily in research, where it pairs its own data with its own compute.
How Bristol Myers Squibb Measures Results
BMS reports targets more than results. Boerner’s goal is to halve the time from target selection to lead molecule. BMS says AI agents save scientists weeks of manual work, a company-claimed figure without a total.
Gaps remain. BMS gives no adoption rate, value figure or cost saving from AI, and it has not linked AI to its $2 billion productivity program. Reporting those links is the step from AI adoption to AI value.
Next markers: progress on the target-to-lead goal, results from Claude agents in clinical and manufacturing work, and headcount for 2026.
Change log
- January 2023: BMS launches its first internal AI chatbot.
- Early 2025: BMS announces a $2 billion strategic productivity initiative through 2027.
- May 2026: BMS makes Claude a shared AI platform for more than 30,000 employees.
- July 2026: BMS announces an NVIDIA AI factory for research.
The Bottom Line
- Disclosure: 4 of 4. BMS answers all four questions, with measurement in part through targets.
- One shared AI layer now spans the company, with Claude for more than 30,000 employees.
- Predict First puts AI before bench work in research.
- Level 3 is evidenced in research, on company-claimed results.
- The say-do gap is results. BMS says AI is paying off but publishes no figures.
- Watch the target-to-lead goal. The test is whether BMS reports that discovery time has been cut in half.
Revenue, Headcount and Revenue per Employee
What the numbers show: Revenue rose 4% from 2022 to 2025 while headcount fell 5%, so revenue per employee rose 10%, to about $1.48 million. Headcount fell by about 1,600 in 2025 as BMS cut costs under two productivity programs. These figures do not show that AI caused any change.
Revenue, headcount and revenue per employee, FY2022 to FY2025
- FY2022: $46.16bn revenue, 34,300 employees, $1.35M per employee
- FY2023: $45.01bn revenue, 34,100 employees, $1.32M per employee
- FY2024: $48.30bn revenue, 34,100 employees, $1.42M per employee
- FY2025: $48.19bn revenue, 32,500 employees, $1.48M per employee
Change, FY2022 to FY2025: revenue up 4%, headcount down 5%, revenue per employee up 10%.
Notes: Revenue is total revenues, as reported. Headcount is approximate, at December 31. BMS is running two cost programs: $1.5 billion announced in 2024 and a $2 billion strategic productivity initiative through 2027.
These figures give context on scale and productivity. They do not show that AI caused any change. Sources: Bristol Myers Squibb 10-K filings.
How Bristol Myers Squibb Compares with US Pharma Peers
We looked at how Bristol Myers Squibb compares with other US pharmaceutical companies in the index. Here is how they stack up.
| Company | Disclosure | AI owner | Fluency floor | Governance | AI value reported | Revenue per employee |
|---|---|---|---|---|---|---|
| AbbVie | 4 of 4 | VP of AI Strategy and Partnerships | Not published; training offered at all levels | Human expertise and ethical use; no published framework | One result (regulatory writing), no total | $1.07M |
| Amgen | 3 of 4 | EVP of research, AI and data; SVP and CTO | Not published; ChatGPT Enterprise across the company | Published AI Vision; phased rollouts | None published | $1.17M |
| Bristol Myers Squibb | 4 of 4 | EVP and Chief Digital and Technology Officer | Not published; frontier models for all staff | Enterprise governance and audit controls; humans decide | Targets set, no results | $1.48M |
| Eli Lilly | 3 of 4 | Chief AI Officer | Level 1, by expectation | Eight principles, board briefings | No: capacity and deals | $1.30M |
| Gilead Sciences | 2 of 4 | Global CIO; no AI leader named | Not published; AI coach in pilot | Six published AI principles | None published | $1.73M |
| Johnson & Johnson | 4 of 4 | EVP and CIO; each function governs its own AI | Level 1, training required before access (2024) | Function-led governance; central AI board removed | Results by use case, no total | $682k |
| Merck | 4 of 4 | CEO sponsors five AI programs; CIO leads technology | Not published; 80%+ use the AI platform | Human review of AI drafts; no published framework | Results by use case, no total | $867k |
| Pfizer | 3 of 4 | Chief AI, Data and Analytics Officer | Level 1, 98% trained | Three principles, a board risk topic | Inside wider cost programs | $834k |
| Moderna AI Pacesetter | 2 of 4 | Chief People and Digital Technology Officer | Level 1, by expectation | Mentioned, not described | No: adoption figures | $414k |
Disclosure counts how many of the four strategy questions (where it is going, where it will focus, what it will take and how it will know it is working) a company answers in public. Fluency floor is the lowest level of AI fluency a company expects of all its people, on the five levels of the Lead with AI Fluency Matrix: Avoiding (0), Chatting (1), Collaborating (2), Integrating (3) and Transforming (4). The other columns summarize each company’s profile. Revenue per employee is for each company’s latest fiscal year, in US dollars. Rows marked AI Pacesetter or Notable are companies outside the S&P 100 and STOXX Europe 50. They are shown for contrast and are not counted in any index figure. See all companies in the AI Strategy Index.
Frequently Asked Questions
What is Bristol Myers Squibb’s AI strategy for its own work?
BMS is building one shared AI platform across the company and putting AI predictions first in research. It made Claude available to more than 30,000 employees in 2026.
Who leads AI at Bristol Myers Squibb?
Greg Meyers, Executive Vice President and Chief Digital and Technology Officer, leads AI across the company, under Board Chair and CEO Christopher Boerner. Chief Research Officer Robert Plenge leads AI in research.
Which AI tools do BMS employees use?
BMS employees use Claude from Anthropic as a shared platform, alongside other frontier models through an internal platform. Engineers and data scientists use Claude Code.
Is Bristol Myers Squibb cutting jobs because of AI?
BMS has not linked job cuts to AI. It is cutting costs under two productivity programs, and headcount fell to about 32,500 at the end of 2025.
How is this profile put together?
Lead with AI profiles use public sources only, such as annual reports, earnings materials, filings and on-record interviews, dated at the time of writing. AI fluency is coded against the Lead with AI Fluency Matrix, and company figures are marked as company-claimed unless they have been independently verified.
About the AI Strategy Index
The Lead with AI AI Strategy Index profiles how the largest companies in the US and Europe approach AI, starting with the S&P 100 and the STOXX Europe 50. Each profile answers four questions: where the company is going with AI, where it will focus, what it will take and how it will know it is working. AI Pacesetters and notable companies outside the two indexes are profiled for contrast and are not counted. Profiles use public sources only, and AI fluency is coded against the Lead with AI Fluency Matrix.
See all company profiles