Google’s AI Generates Novel Cancer Therapy Hypothesis, Validated in Human Cells

Google DeepMind's latest biological AI system has achieved a scientific milestone by generating and experimentally confirming a new hypothesis for cancer treatment. The Gemma-based model, developed in collaboration with Yale University, identified silmitasertib as a conditional amplifier that enhances tumor visibility to the immune system.

The 27-billion-parameter foundation model, called Cell2Sentence-Scale 27B (C2S-Scale), represents a leap forward in AI's capacity for scientific discovery. "This discovery may reveal a promising new pathway for developing therapies to fight cancer," said Google CEO Sundar Pichai. The breakthrough suggests scaled biological AI models can invent—not just optimize—novel scientific concepts.

While preclinical and clinical tests remain, the validation in living human cells marks a significant step toward AI-driven drug discovery. The achievement underscores the growing convergence of artificial intelligence and life sciences, with potential implications for biotechnology investment trends.