GSK has entered a five-year strategic collaboration with San Francisco-based AI-native biotech Noetik, investing $50 million upfront plus near-term milestones (potentially reaching $888 million total) for non-exclusive access to Noetik’s OCTO-VC virtual cell foundation models in non-small cell lung cancer (NSCLC) and colorectal cancer (CRC). This “model licensing” paradigm combines GSK’s immunology expertise with Noetik’s generative AI for virtual tumor simulations, accelerating target identification and patient stratification.
Glimpse:
The deal grants GSK subscription-based access to Noetik’s self-supervised AI models trained on vast spatial multimodal data for simulating gene expression, cell states, and tumor-immune interactions. It validates a new biotech asset class: licensing human foundation models rather than candidates. Noetik’s platform enables “what if” biological queries beyond real-patient limits, targeting challenging cancers with novel, deterministic drug engineering.
GSK has forged a transformative partnership with Noetik, paying $50 million upfront and committing to near-term milestones in a deal potentially worth up to $888 million. Announced January 8, 2026, the five-year agreement provides GSK non-exclusive licenses to Noetik’s OCTO-VC virtual cell models for NSCLC and colorectal cancer two notoriously difficult tumor types.
Noetik’s AI-native approach generates high-quality spatial data at scale via self-supervised learning, training foundation models on multimodal human datasets. These “virtual cells” simulate complex biology: gene expression, cellular states, tumor-immune dynamics enabling in silico hypothesis testing that surpasses limited real-world samples.
Key Deal Elements:
Upfront & Milestones: $50 million initial + near-term payments.
Access Model: Annual licensing fees for ongoing model use and refinements.
Collaboration: Joint efforts blending GSK’s AI/tumor immunology leadership with Noetik’s simulation tech.
Focus: Sharpen target selection, patient stratification, and novel medicine development.
Executive Insights: Kim Branson, Ph.D., GSK Global Head of AI/ML: “Noetik’s approach to generating high-quality spatial data at scale to train foundation models is novel. It has the potential to deepen our understanding of biology and support development of novel medicines.”
Ron Alfa, M.D., Ph.D., Noetik CEO & Co-Founder: “This agreement validates a new paradigm in biotech: the licensing of human foundation models… GSK is now equipped with one of the most extensive oncology multimodal spatial training sets in existence.”
Daniel Bear, Ph.D., Noetik VP of AI: “These models let us go beyond limited patient data to ask ‘What if?’ questions about genes, proteins, cells, and tissue.”
This deal following Noetik’s prior collaborations establishes AI model licensing as a viable biotech revenue stream, shifting from service fees to infrastructure assets.
“We built Noetik to move the industry from probabilistic ‘shots on goal’ to deterministic engineering of cancer drugs.”
By
HB Team
