Hill Research Co-Hosts Clinical AI Mastermind S1 at J&J Innovative Medicine
On May 20, 2026, Hill Research co-hosted Session 1 of the Clinical AI Mastermind 2026 at Johnson & Johnson Innovative Medicine in Cambridge. The topic: The Submission Last-Mile: Redefining Clinical Velocity.
The Mastermind is a bimonthly, curated forum for senior pharma, biotech, regulatory, and clinical AI leaders to compare notes on what’s actually working — not vendor pitches, not polished demos. Session 1 was the first of four planned across the year.
Hosts
- Anchor host: Johnson & Johnson Innovative Medicine
- Co-hosts: Hill Research · AKT Health · Tech Impact Foundation
The Curator’s Statement
“The Last-Mile is no longer primarily a data problem. It is an architecture problem.”
The conversation was framed around three pillars: Scale, Agility, and Integrity by Design.
Panel
- Moderator — Dr. Charmaine Demanuele, VP & Head of R&D Data Science & Digital Health for Neuroscience, J&J Innovative Medicine
- Scale — Dr. Rogier Landman, Associate Director, Digital Medicine Data Science, Pfizer
- Integrity — Dr. Naveed Afzal, Head of Data Science, Takeda
- Agility — Bhaskar Dutta, Alexion Pharmaceuticals
- Co-producer — David Hall, AKT Health
- Co-host remarks — Dr. Alexander Duprey, Chief Product Officer, Hill Research
Hill Research’s Take
Hill Research Chief Product Officer Dr. Alexander Duprey opened the co-host remarks with the line that became the room’s working hypothesis:
“Data is plentiful. Analyses are plentiful. The limitation now is operationalization.”
The paradox: every organization has more data, more systems, more dashboards, more AI pilots than ever — yet the processes used at submission stage are largely the same ones built 30 years ago. Closing the gap between clinical intent, statistical rigor, regulatory expectation, and AI is the unsolved problem Hill Research is tackling head-on with TriClick.
Lines That Stuck
Dr. Naveed Afzal (Takeda) — on integrity:
“Compliance and governance aren’t slowing us down. They’re what lets us move quickly in the Last-Mile.”
“67% of CFOs say AI is underperforming. The majority of the time the reason is the same — reproducibility wasn’t designed in from the start.”
“Treat AI outputs as evidence objects, constructed continuously — not as insights at the end.”
Dr. Rogier Landman (Pfizer) — on scale:
“We won’t have human-out-of-the-loop solutions. And you don’t have to do everything with AI — use it for the parts AI is good at, not the whole thing.”
On smaller biotechs: “They have one major advantage — they can architect their data for AI from day one. Large pharma is still untangling decades of legacy.”
Bhaskar Dutta (Alexion) — on trust:
“The biggest barrier to scaling isn’t the technology — it’s trust. Earn it once on a peer-reviewed methodology, and it snowballs. Programs come to you.”
“QC may be one of the best use cases for AI right now — multiple agents cross-checking each other’s work before a human ever sees it.”
Dr. Charmaine Demanuele (J&J) — on culture:
“Every new technology that makes a big impact requires a culture shift, not just a technology shift.”
On the FDA’s real-time clinical trials pilot: “This isn’t really a policy change — it’s a change in operating model.”
The Underlying Theme
Across all three pillars — scale, agility, integrity — the Last-Mile problem is no longer about producing the analyses. It’s about constructing a system where every analysis arrives with its provenance, lineage, version, audit workflow, and accountability built in. The architecture is the work.
Thanks
Thank you to Dr. Demanuele, Fiona Elwood, Faith Loller, Angelica Weizman, and the J&J Innovative Medicine team for hosting; to David Hall for co-producing; to Bhaskar Dutta for stepping in on short notice; and to every Director+ guest who chose this conversation over an inbox of email.
Session 2 of the Clinical AI Mastermind is bimonthly. We’ll share the topic and venue soon.
Learn More
- For partnership inquiries, contact: info@hillresearch.ai