This Week on Life Sciences Digital
Roche activated 2,176 new NVIDIA Blackwell GPUs this week, bringing its total to over 3,500 across the U.S. and Europe — the largest declared AI compute footprint in the pharmaceutical sector. The architecture is built around the Lab-in-the-Loop framework: continuous feedback between wet-lab experiments and AI models, with NVIDIA Omniverse powering digital twins of production lines and BioNeMo handling biology-specific model training. This is not a research unit. It is a new operating model for drug development across discovery, manufacturing, and diagnostics.
Roche's partnership with NVIDIA, which started with Genentech in 2023, now covers the entire organization. Eli Lilly announced a concurrent $1 billion AI commitment with NVIDIA as its partner. The top tier of pharma is converging on a single compute stack. The decisions being made now will be very hard to reverse.
Visual summary generated with AI (NotebookLM).
Also This Week:
At GTC 2026, IQVIA announced IQVIA.ai: a unified agentic AI platform built with NVIDIA, already adopted by 19 of the top 20 pharmaceutical companies. It embeds autonomous AI directly into clinical, commercial, and real-world evidence workflows, learning continuously from operational feedback. IQVIA is the world's largest CRO. Its platform choices become the de facto standard for the industry that runs on its infrastructure.
Earendil Labs closed a $787 million round with Dimension Capital, DST Global, and Sanofi. The platform has generated over 40 therapeutic programs, with one already in Phase 2. Sanofi holds exclusive licensing rights for next-generation bispecific antibodies coming out of the platform. This is a new financing template for AI drug discovery: prove program-level output, partner with a top-10 pharma, raise at execution scale before an acquisition is necessary.
Google and DocMorris — Europe's largest online pharmacy — announced a full-stack AI partnership across Germany, Switzerland, and Austria. DocMorris is migrating its entire IT infrastructure to Google Cloud, deploying Gemini-powered conversational AI across the prescription workflow, and building a personalized health companion from symptom assessment to e-prescription fulfillment. Google's U.S. playbook — embed at the infrastructure level before competitors can — is now running in Europe.
Signals & Market Moves
Verily is rebranding as Verily Health Inc. and converting to a corporation, with Alphabet now holding a minority stake. The $300M round came with three strategic partners: NVIDIA for the Pre clinical research platform, Samsung Galaxy Watch for biomarker development, and Salesforce Agentforce Health for enterprise deployment. The closure of Verily's medical device division last year was the last signal of what it is becoming: an AI-native precision health software platform, operating independently.
The Signal: NVIDIA + Samsung + Salesforce in a single round describes an integration layer, not a standalone product. Verily is positioning itself as connective tissue between health data, AI compute, and enterprise workflow — not as a device or app company. Worth watching which health systems and pharma companies land on its platform over the next 12 months.
A Federal Court Just Made Shadow AI a Legal Risk, Not Just an IT Problem 🔗
In United States v. Heppner, a New York federal judge ruled that documents prepared using public generative AI platforms do not qualify for attorney-client privilege — because there is no reasonable expectation of confidentiality on third-party AI tools. For life sciences companies routinely passing regulatory submissions and clinical data through AI models, this creates direct exposure.The Signal: This ruling landed the same week that Pharmaceutical Executive published a governance framework piece on shadow AI in pharma. Two independent sources, one legal and one operational, naming the same risk in the same week. Platforms with built-in audit trails and data sovereignty — IQVIA.ai, Rivia, DATAN Analytics AI+ — now have a compliance argument that general-purpose AI tools cannot make. Governed AI is no longer just best practice. It has precedent behind it.
PharmaMar and Globant deployed a multi-agent AI framework for oncology drug discovery: 20+ specialized agents across preclinical, clinical, regulatory, and strategic domains, processing 4,500 cancer documents and identifying the ten most viable treatment candidates from over 8,000 options — at 15 times the speed and over 90% accuracy. Next phase adds automated hypothesis generation and compliance checking.
The Signal: This is what production-grade agentic AI in pharma actually looks like — not one general model, but domain-specific agents in a governed environment. It is structurally what IQVIA.ai is building as a commercial service. Both paths lead to the same place: AI in the execution loop, humans in the review loop.
Tool Spotlight from our Life Sciences Digital database
AI agents built specifically for clinical development workflows — automating statistical analysis plans and SAS code generation within ISO 27001-certified, GDPR-compliant infrastructure. Designed for CROs and pharma sponsors running CDISC-compliant submissions.
Got a tool for life sciences you’d like more people to know?
Events & Calls
BIO-Europe Spring 2026 — Lisbon, March 23–25 Europe's largest biotech partnering conference opens today, with over 3,700 professionals from 60+ countries. Therapeutic Insights and Ecosystem Innovation are the tracks to follow.
Evidence for AI in Health (EVAH) — $60 Million Open Call Deadline: April 1, 2026 Wellcome Trust, Gates Foundation, and Novo Nordisk Foundation. Pathway A: up to $1M. Pathway B: up to $3M. Focus: AI decision support tools for frontline health workers in LMICs.
Impact Challenge: AI for Science — Google.org Deadline: April 17, 2026 $500K–$3M grants plus six months of pro bono Google technical support and Cloud credits. Focus areas include AI for Health & Life Sciences. One of the most accessible funding mechanisms in the market for research-stage organizations and nonprofits.
🤝 Got a tool for life sciences you'd like more people to know about?
Submit your tool here.
Want to be featured in a future issue or explore sponsorship?
We highlight the AI tools, companies, and initiatives shaping the future of life sciences. If you'd like to collaborate, reply to this email or get in touch: [email protected]
