Cellular Intelligence Adds LeCun and Langer to Its Scientific Advisory Board

A man who won the Turing Award for deep learning and a man who founded Moderna now appear on the same advisory board roster. Cellular Intelligence, an AI company based in Boston, announced on September 21 that Yann LeCun, Robert "Bob" Langer, Jens Nielsen and Fabian Theis have joined its Scientific Advisory Board. Langer is also joining the board of directors as an observer. A company that intends to build a universal foundation model of cell signaling has seated the top name from the AI side and the top name from the medical side at one table.

According to the announcement, the four bring between them a Turing Award, more than 1,500 issued and pending patents, the founding of Moderna, the leadership of the Novo Nordisk Foundation's BioInnovation Institute, and Germany's highest research honor. The company wrote that their work touches directly on the two problems it was founded to solve: learning predictive models from complex systems, and carrying that understanding into therapeutics.

Yann LeCun received the 2018 ACM A.M. Turing Award for foundational work in deep learning, founded Meta's Fundamental AI Research lab, FAIR, and served as the company's chief AI scientist from 2013 to 2025. He is now executive chairman of AMI Labs, the world-model company he co-founded, and a Silver Professor at New York University. In the announcement he said, "The models that interest me learn how the world works from data and predict what happens when we act on it," adding that "living cells present an extraordinarily rich version of that challenge." METAL has reported on world-model research being gathered into a Royal Society theme issue, and LeCun's choice of the cell pulls that current into biology.

Bob Langer is an Institute Professor at MIT, the school's highest faculty rank, and the most cited engineer in history. He holds more than 1,500 issued and pending patents, has co-founded Moderna and roughly 40 other companies, and has received the National Medal of Science, the National Medal of Technology and Innovation and the Queen Elizabeth Prize for Engineering. In the announcement he said, "What draws me to this company is the opportunity to connect a fundamental understanding of cell behavior with the practical work of developing medicines," and added, "If we can better predict how cells respond to an intervention, we can ask more precise questions, design better experiments and open new possibilities for patients."

The remaining two seats went to European computational biology. Jens Nielsen is chief executive of the BioInnovation Institute, the Novo Nordisk Foundation's Copenhagen engine for founding and scaling life-science companies, funded at more than three billion Danish kroner, and a professor of systems biology at Chalmers University of Technology. He has written more than 900 scientific publications and was knighted by Queen Margrethe II of Denmark in 2019. Fabian Theis directs the Computational Health Center at Helmholtz Munich, is scientific director of Helmholtz AI, and co-founded the Human Cell Atlas, an international project to map every cell type in the human body. He received the 2023 Gottfried Wilhelm Leibniz Prize for machine-learning methods that predict how cells develop and respond to drugs.

Micha Breakstone, co-founder and chief executive of Cellular Intelligence, explained what the company is building using the phrase world model. According to reports, he said, "A world model learns how a system behaves and predicts what happens when you act on it," and added, "For us, the system is a living cell: the basic computational unit of life, receiving signals, processing information and changing its behavior in response." The idea is to carry the method that predicted text and images straight over to the cell.

These appointments came after the company already had a clinical asset in hand. In May 2026 Cellular Intelligence acquired global rights from Novo Nordisk to STEM-PD, a cell therapy for Parkinson's disease. It is an allogeneic therapy designed to replace the dopamine-producing neurons lost to Parkinson's, and it holds US Food and Drug Administration Fast Track designation and investigational new drug clearance, leaving it ready to enter Phase 2. Novo Nordisk had shelved the program when it shut down its cell-therapy division last October, and in the deal the company took an equity stake in Cellular Intelligence while remaining eligible for future milestones and royalties.

The company changed its name in January. The company news list METAL reviewed shows that Somite, founded in 2023, became Cellular Intelligence on January 7, 2026, saying it would build a universal virtual cell-signaling model; before that, in May 2025, it raised a Series A of more than $47 million led by Khosla Ventures, and in August 2025 it took a strategic investment from AMD Ventures. It has now raised more than $70 million, and its investors include Khosla Ventures, the Chan Zuckerberg Initiative and Novo Nordisk. On August 12 the single-cell biology researcher Arjun Raj joined as chief scientific officer and scientific co-founder.

The scorecard the company offers is still an internal one. According to reports, Langer said early internal benchmarks showed stronger prediction results than existing models in both healthy development and cancer, and that this happened "despite never training on a single cancer cell." Breakstone also said that Fidji Simo, chief executive of applications at OpenAI, had helped shape the company's direction as a strategic adviser. STEM-PD remains the only clinical program the company owns, and it says it will also build a business licensing its models to pharmaceutical companies for paid drug-discovery work.

Advisory appointments usually read as publicity, but this particular roster reveals something else. Seating an AI researcher with no biology background next to a medical researcher with no AI background shows, in reverse, that the company has not yet been fully credentialed by either academic community. Computational biology grew up with two communities that do not know each other's review standards, so trust crosses over through people rather than papers. The moment a Turing Award and a National Medal of Science appear on the same line, both fields have a reason to look at this company once more.

The value of the model will ultimately be set in front of patients. METAL has reported that AI-designed proteins diverged between prediction and actual results. A cell-signaling model will be priced in the same place, and the fact that this company holds a clinical program of its own means it intends to run that comparison with its own hands. The announcement said manufacturing and clinical data from STEM-PD flow back into the company's models, and it also said the therapy is investigational, with safety and efficacy not yet established. Whether the predictions hold will show up in patients, not in a roster.