👋 Good morning! Once upon a time, a mild headache sent you to WebMD, where five clicks later you were emotionally preparing to be diagnosed with a rare disease last seen in a remote jungle in sub-Saharan Africa.

Now, OpenAI would like a word.

Last week, OpenAI rolled out ChatGPT Health, a dedicated space for users to ask health and wellness questions. According to the company, more than 230 million people already ask ChatGPT health-related questions every week. This is WebMD scale behavior, just with better grammar and way more confidence.

The pitch is familiar. Healthcare is expensive. Doctors are booked. Continuity of care is broken. AI can help fill the gaps. But there’s a catch OpenAI itself acknowledges. Large language models do not know what is true. They predict what sounds right. And hallucinations are not exactly what you want when discussing chest pain.

OpenAI is clear that ChatGPT Health is not for diagnosis or treatment. But WebMD taught us one thing. Give people partial medical information and they will sprint directly to worst-case conclusions and self diagnosis. Now they can do it conversationally… Progress.

🤝 Eli Lilly and Nvidia Commit $1B to an AI Drug Discovery Lab
Eli Lilly and Nvidia are deepening their partnership with a new AI co-innovation lab backed by more than $1 billion over five years. The Bay Area facility will pair Lilly’s drug development infrastructure with Nvidia’s AI stack, including foundation models and DGX Cloud compute. The goal is closed-loop discovery that moves faster from hypothesis to clinic. This is not a pilot. It is a statement that compute is now a core R&D input.

💸 AbbVie Pays $650M Upfront to Enter the PD-1xVEGF Arms Race
AbbVie handed RemeGen $650 million upfront for ex-China rights to RC148, a PD-1xVEGF bispecific antibody, with total deal value reaching $5.6 billion if milestones hit. The asset could pair neatly with AbbVie’s ADC portfolio across lung and colorectal cancers. In parallel, AbbVie announced a $175 million investment to acquire and expand a drug-device manufacturing site in Arizona. The message is clear. Pipeline and production are being built together.

🧠 Novartis Spends $165M Upfront on an Alzheimer’s Brain Shuttle
Novartis struck a $1.7 billion deal with SciNeuro to license an antibody platform designed to ferry Alzheimer’s drugs across the blood-brain barrier. The upfront payment clocks in at $165 million, with $1.5 billion in milestones behind it. Bypassing the blood-brain barrier remains one of pharma’s hardest problems. Novartis is betting that transport, not just targets, is the unlock.

😳 Revolution Medicines Becomes a $30B Takeout Target Overnight
Buyout rumors sent Revolution Medicines soaring after reports that Merck is in talks to acquire the RAS-focused oncology biotech for $28 billion to $32 billion. AbbVie was also named as a suitor before publicly denying interest. With a market cap just over $20 billion, any deal would be a massive premium. Whoever lands Revolution gets a deep RAS pipeline and instant credibility in one of oncology’s most competitive battlegrounds.

📈 Aktis Oncology Prices a $318M IPO and Cracks the Public Market Door
Aktis upsized its IPO to 17.6 million shares at $18, raising roughly $318 million and becoming the largest biotech IPO since the market cooled. That figure beats every biotech offering from last year and signals real appetite for late-stage oncology stories. Two more biotechs, Veradermics and Eikon Therapeutics, are already lining up behind it. The IPO window may finally be open, at least a crack.

🦾 Eli Lilly Adds Another AI-Powered Oncology Deal With InduPro
Fresh off other AI collaborations, Lilly signed a deal with InduPro Therapeutics worth up to $950 million to pursue bispecific and multispecific cancer targets. InduPro will lead early discovery using its AI-enabled membrane interactomics platform. Lilly continues to treat AI not as a vendor tool but as a pipeline multiplier.

💵 Teva Secures $500M to Push a Vitiligo Program Forward.
Royalty Pharma will bankroll development of Teva’s anti-IL-15 antibody with an initial $75M tranche funding a phase 2b trial this year.

🎯 Roche Commits $570M to Another ADC Outside China.
MediLink’s B7-H3-targeting ADC is already in phase 2 trials, and Roche wants in everywhere but China.

🏛️ Johnson and Johnson Signs a Most-Favored-Nation Pricing Deal.
The agreement unlocks tariff relief and includes new U.S. manufacturing investments in Pennsylvania and North Carolina.

🧬 FDA Loosens Manufacturing Rules for Cell and Gene Therapies.
Regulators are carving out flexibility for personalized and small-batch CGTs to speed development.

🐳 Orca Bio Raises $250M Ahead of a Pivotal FDA Decision.
The cash will support commercialization prep for its allogeneic T-cell therapy pending an April verdict.

🧪 EpiBiologics Closes a $107M Series B Backed by J&J and Google.
The protein degradation startup plans to enter the clinic later this year with its lead EGFR degrader.

Pharma executives love talking about the future. PwC just handed them a glossy roadmap to 2035 filled with AI, patient-centric everything, and bold new R&D frontiers. The catch is that most of these ambitions require doing the exact opposite of what investors have rewarded for the past decade.

PwC’s prescription for 2026 is aggressive. Embed AI everywhere. Reinvent R&D. Move beyond crowded therapeutic categories. Chase big, unsolved problems like organ regeneration and lifespan extension. It sounds inspiring… and expensive. Conveniently, we just watched Eli Lilly and Nvidia put real money behind that vision with a billion-dollar AI lab.

But here is the tension. Investors consistently punish companies that step too far outside proven lanes. Moonshots are applauded in theory and shorted in practice. Safe, incremental programs still dominate capital flows. If pharma truly pivots toward bolder science, will capital follow or flee? 🤷‍♂️

A parallel report on Massachusetts biopharma shows the contradiction clearly. The state’s pipeline grew nearly 14 percent in 2025 even as venture funding hit a six-year low. Innovation is accelerating. Money is not. China’s pipeline growth, meanwhile, jumped 37 percent in the same period.

The data suggests the current model is strained. Pricing pressure, regulatory uncertainty, and rising competition are eroding old playbooks. AI feels like the safe bet because it improves efficiency without challenging investor comfort. The risk is that everyone crowds into tools and no one funds the breakthroughs those tools are meant to enable.

If biopharma needs reinvention, capital will have to change too. Otherwise, we are just automating the same old decisions faster.

This is the latest sales figure for Insmed’s newly approved bronchiectasis drug, Brinsupri. It delivered a blowout first full quarter, nearly tripling Wall Street expectations and signaling a rare clean launch win.

Here’s what biotech Redditors are talking about:

😕 Why is Boston Biotech Stuck in Neutral
One redditor asked why jobs have not rebounded despite capital markets showing signs of life. The consensus was blunt. Companies are running leaner, excess lab space from the COVID boom still needs to be absorbed, and headcount is no longer the growth signal it once was. Several commenters expressed frustration that executive pay cuts are never part of the correction conversation. The good old days of social hours and expense approvals are remembered fondly, and quietly mourned.

💁‍♀️ Do You Actually Need to Vibe With Your Manager?
Another thread debated whether a lack of personal rapport with a manager is a real problem. Many argued it is not. Professional, respectful, and predictable beats forced friendliness every time. Others blamed remote work for stripping away organic connection. A few admitted they prefer managers who do not ask about their weekend. Different strokes. Same Slack channel.

🤖 Utah Lets AI Renew Prescriptions Without a Doctor.
The state launched a pilot allowing AI to handle routine refills for chronic conditions. What could go wrong? 👀

🤝 Pfizer Partners With Boltz to Build Custom Biomolecular AI Models.
The collaboration will produce exclusive structure and affinity prediction tools.

💰 GSK Commits $50M to Noetiks’ Cancer AI Models.
The deal focuses on spatial biology and patient-specific efficacy prediction.

📺 FDA Slaps Esperion Over Its Animated Cholesterol Ads.
Regulators say the “Lipid Lurkers” are distracting viewers from real risk information.

❤️ Lexeo Teams With J&J’s Abiomed on Gene Therapy Delivery.
The plan uses heart pump tech to localize AAV delivery directly to cardiac tissue.

🧬 Aurora Therapeutics Wants to Make One-Patient Drugs Scalable
Backed by Menlo Ventures, CRISPR pioneers Jennifer Doudna and Fyodor Urnov launched Aurora to tackle ultra-rare diseases, including conditions affecting a single patient. The thesis is simple and radical. The tools finally exist. Now the system needs to catch up.

🚗 January 13, 1942 — Henry Ford patented the Soybean Car, a lightweight vehicle built from agricultural plastics made of soybeans, wheat, and corn. Bio-based materials were cool before it was cool.

That’s all for today. JPM Healthcare Week is just getting started in San Francisco, which means this calm will not last.

If history holds, Thursday’s newsletter will be fueled by deal rumors, AI demos, and at least one slide deck claiming to reinvent medicine.

Until then, may your models converge and your diagnoses stay off ChatGPT Health.

Was this email forwarded to you? Sign up here 👉 Subscribe for free

Follow us on social and stay one step ahead