NVIDIA AI Podcast – Episode 264
How Bristol Myers Squibb Is Accelerating Drug Development With AI
Host: Noah Kravitz
Guest: Greg Meyers, EVP & Chief Digital and Technology Officer, Bristol Myers Squibb
Date: July 9, 2025
Episode Overview
In this episode, host Noah Kravitz sits down with Greg Meyers of Bristol Myers Squibb (BMS) to explore how one of the world’s leading biopharmaceutical companies is harnessing artificial intelligence (AI) to transform drug discovery, accelerate development timelines, and reshape life sciences research. Meyers discusses the company’s AI-driven initiatives—ranging from molecular prediction to empowering employees with generative AI tools—and explores how these advances are moving the entire biopharma industry toward more personalized and effective therapies.
Key Discussion Points & Insights
1. Bristol Myers Squibb’s Mission and Digital Transformation
- Company Focus: BMS is dedicated to developing innovative medicines for patients with serious diseases, including cancer, cardiovascular conditions, immunology, and neuroscience (01:24).
- Ambitious Goals: BMS aims to deliver 10 new medicines and 30 new indications by 2030, a vision powered by enterprise-wide integration of digital, data, and AI (01:39).
- Greg Meyers’ Role: Oversees technology, data & analytics, and digital health, with a focus on modernizing patient diagnosis, treatment, and monitoring (02:04).
2. The Evolution of Drug Development with AI
- Historical Challenge: Traditional drug discovery leaned heavily on human intuition and “hunches” at every step, from identifying disease patterns to developing testable hypotheses (04:26).
- Quote: “If you look at every, probably most people learned at some point in school the scientific method and the first three steps, if you were to summarize them, are more or less follow your gut.” — Greg Meyers (04:26)
- AI’s Role: AI now narrows down these hunches by:
- Uncovering hidden patterns in biological data invisible to humans
- Accelerating the analysis of cancer tumors and diagnostics via convolutional neural nets
- First-Generation AI Molecules: The emergence of “co-designed” molecules created with AI guidance, already advancing into human trials at BMS (06:56).
3. Concrete AI Applications in Research & Development
- Case Study — Sickle Cell Disease (06:57):
- AI helped design a protein to degrade another protein that suppresses a beneficial gene in sickle cell patients, aiming to alleviate disease symptoms by restoring infant-type hemoglobin.
- Quote: “What AI has helped us to do is to sort of engineer features of that molecule that allowed us to be able to effectively turn back on the dormant genes you have as infants.” — Greg Meyers (07:26)
- Quantifying AI Impact:
- Predictive molecule invention now drives experimentation, relying on millions of historical compounds and experiments to build success-likelihood models (08:50).
- In “small molecule” research, 100% of experiments are now filtered by AI-driven predictions before moving to the lab, with similar strategies adopted for “large molecule” projects (09:27).
- AI dramatically accelerates mundane but critical work—e.g., informed consent forms for clinical trials are now 80% pre-written with LLMs (10:45).
4. Empowering Staff with Generative AI Tools
- Internal Adoption:
- BMS rapidly rolled out company-wide tools in the wake of ChatGPT’s release, including a Teams-integrated chatbot and a “Gen AI storefront” offering access to leading models (GPT-4, Anthropic, DeepSeek, Gemini, MS Copilot) for free to employees (12:04).
- “We wanted to build safe alternatives…We didn’t want people going out and leaking proprietary information into consumer tools.” (13:52)
- Cultural Shift:
- Early use centered around document editing, performance reviews, and knowledge retrieval.
- Usage is now expanding to include advanced analytical and data science tasks, bolstered by staff training and safe experimentation (13:52).
5. Custom and Open-Source AI for Scientific Modeling
- Scientific Computing:
- Much computational biology and chemistry runs on NVIDIA clusters (15:18).
- Protein structure modeling uses tools like AlphaFold and open-source LLMs (e.g., ESM Fold).
- Emphasis on building abstraction layers and copilots to make these tools accessible for scientists who aren’t coders (16:54).
- Quote: “A lot of this work is still at the early adopter phase…We’re working on building abstraction layers to create these copilots…to help scientists with really specific problems that they face on a day to day basis.” — Greg Meyers (16:25)
6. AI and the Larger Biopharma Ecosystem
- Data Silos: 30% of the universe’s data is healthcare-related, but it’s locked in incompatible silos across hospitals, companies, insurers, and governments. Integrating these is critical for progress (17:53).
- Personalized Medicine: AI can predict which patient will respond to which therapy, potentially sparing patients from failed treatment cycles, especially in diseases like lung cancer (19:16).
- Future Requirement: In two decades, shipping a drug “alone won’t be enough”—bundling real-world evidence and patient data will be standard (19:44).
7. Overcoming Healthcare Data Fragmentation
- Electronic Medical Records (EMRs):
- Diagnostic Advances:
- New AI tools now analyze routine ECGs to screen for diseases like hypertrophic cardiomyopathy, catching cases that previously went undetected for years (23:25).
- Quote: “What we were working on with this partner is being able to create the ability to look at a simple 12 lead ECG…the ability to actually find, to detect the signature of that disease…really unlocks the potential to uncover people who didn’t know they have the disease.” — Greg Meyers (23:25)
- AI can re-examine legacy medical images/data to discover overlooked patterns (25:12).
8. The Coming Decade: Transformational Acceleration
- Clinical Trials: BMS is on track to shave 2–3 years off traditional 10-year drug development timelines thanks to AI, data, and process changes (27:19).
- Quote: “We’re on track to shave off almost three years off of our clinical trial timeline as a result of using digital and data and AI and process changes.” — Greg Meyers (27:19)
- For patients with few options, this is “transformational.”
- Therapeutic Frontiers: Next-wave cancer treatments—cell therapies, antibody-drug conjugates, radioligands—stand to benefit greatly from accelerated pipelines.
9. Staying Ahead in Rapidly Evolving Tech Landscapes
- Personal Tips: Meyers relies on Reddit and podcasts for fast news, journal articles (sometimes converted to audio via NotebookLM) for depth; believes tech executives must focus as much on external signals as internal work (28:38).
Notable Quotes & Timestamps
-
“In each [cell] is about 1 trillion molecules. If you add that up, that’s 7 octillion atoms. Any one of those atoms out of place could cause, cure, or prevent a disease.”
— Greg Meyers (03:01) -
“In about 100%…small molecule discoveries…we would not run experiments in the wet lab until those predictive models…suggest it would be worth trying.”
— Greg Meyers (09:27) -
“We wanted to build safe alternatives…we didn’t want people going out and leaking proprietary information into consumer tools.”
— Greg Meyers (13:52) -
“The ability to get something to patients two to three years earlier can be transformational.”
— Greg Meyers (27:24) -
“I think…in 20 years…just shipping a molecule alone won’t be enough. We’re going to need to…bundle together real world evidence about how medications work…”
— Greg Meyers (19:44)
Key Timestamps
- [01:24] — Introduction to Bristol Myers Squibb and Meyers’ role
- [04:26] — How scientific discovery has historically relied on human “hunches”
- [06:56] — Specific case: AI-engineered treatment for sickle cell disease
- [09:27] — Quantitative leap: AI-driven experiment prediction and decision-making
- [12:04] — BMS’s internal adoption of generative AI tools
- [15:18] — Use of NVIDIA clusters and open-source models for computational biology/chemistry
- [17:53] — Data fragmentation in the biopharma ecosystem
- [23:25] — AI diagnostic tools for real-world disease detection (e.g., hypertrophic cardiomyopathy)
- [27:19] — Three years cut off clinical trial process via AI
- [28:38] — Meyers on staying informed amid rapid innovation
Closing Thoughts
Greg Meyers paints a compelling picture of a biopharma industry on the brink of reinvention. AI and machine learning have moved from buzzwords to essential infrastructure at Bristol Myers Squibb, impacting everything from molecular design and clinical documentation to population-level insights and point-of-care diagnostics. With a clear-eyed view of data challenges, regulatory headwinds, and workforce transformation, Meyers signals that the journey has just begun—but dramatic gains in patient outcomes and therapeutic innovation are within reach.
Further Information:
Visit bms.com for details on Bristol Myers Squibb’s current therapies, pipeline, and technical initiatives.