Small print.
Wide open.

[Embedded graphic 1 — OCR text]
lh Bristol Myers Squibb

Customer since 2023

Speeding up new medicine discovery

Results

Employees with Claude
access (all)

30,000+

Internal data sources connected,
with thousands more under
consideration, via MCP integrations

Dozens

Time to answer a real-world
evidence question

~2mo>
~5 min

Background

Bristol Myers Squibb is one of the world’s largest
pharmaceutical companies. Over decades, it has amassed
one of the deepest scientific archives in the industry,
spanning research, clinical, manufacturing, and commercial
data. BMS wanted to connect its people, systems, and
institutional knowledge to unlock data and expertise long
trapped in the disconnected systems that define biopharma
today—across its entire value chain, from early drug
discovery through manufacturing quality and commercial
operations. The company began using Claude in 2023,
selecting Anthropic’s models for their large context
windows and high-quality reasoning.

Impact

In May 2026, BMS selected Claude Enterprise as its
company-wide AI platform, chosen for its agentic
capabilities, security posture for regulated environments,
and rapid pace of innovation. BMS first rolled out Claude
Code to its global developers and data scientists, now more
than 3,000 strong, freeing them from having to build and
maintain their own orchestration and agent infrastructure
so they could focus on building new products instead.
Along the way, BMS codified its design patterns and
technical stack into Claude Code through Skills and MCP
integrations, improving output consistency and
accelerating developer onboarding. The company is now
exploring how to extend this model to help scientists,
manufacturing teams, and sales representatives ramp up
faster and focus on higher-value work. As Claude Code was
adopted, BMS expanded its rollout and now plans to bring
Claude Cowork to all 30,000+ employees.

BMS is embedding Claude into the work of bringing
medicines from research to patients. Across the company, the

“Most enterprise AI stops at the chatbot. The
real prize is the untapped value still trapped
behind decades of data silos... Anthropic’s
Claude gives us the agentic capabilities, pace
of innovation, and security necessary to
connect our systems.”

Greg Meyers, EVP and Chief Digital and Technology
Officer at Bristol Myers Squibb

goal is to make decades of siloed data instantly accessible and
actionable, no matter where it lives or which function owns it.
Using Claude, researchers can query, work with, and analyze
company data directly instead of relying on dashboards built
by internal teams. Claude now works across the 30 to 40
external medical records and claims databases BMS
subscribes to, cutting the time to generate an answer to a real-
world data question from up to two months to five minutes or
less. Scientists connect lab notebooks and past studies to find
drug discovery opportunities and pursue investigations
faster, across a broader span of data. BMS's teams are already
exploring how Anthropic’s newly announced Claude Science
can help the company move faster across early-stage
discovery. That same spirit of exploration extends beyond the
lab: operations teams are beginning to query HR software
and IT service management systems in plain language,
without relying on engineering teams.

Each model generation widens the scope of work that BMS
entrusts to Claude. With Claude Opus, the company’s
experts are treating Claude as a PhD-level advisor and
thought partner. There are also several new use cases
currently under evaluation. Looking ahead, BMS is
considering how it might apply Claude’s most powerful
available models to explore autonomous task work with a
human in the loop. For example, an employee could hand
Claude a task—resolving an HR or IT request that spans
several systems, or connecting lab notebooks and past
studies to find new avenues for drug discovery—with
Claude reasoning through which internal systems to use
and how to structure the work, drawing on the company’s
library of MCP integrations.