AI isn’t close to curing cancer. This startup says it knows what it will take.
AI-summarised brief · reviewed before publication
Biotech startup Vivodyne argues that AI drug discovery is hindered by a lack of causal biological data, relying instead on animal testing or single-cell studies. To address this, the company developed HIVE, modular robotic labs that autonomously grow and dose twenty types of human tissue. CEO Andrei Georgescu claims these systems generate the complex, living-tissue data current AI models lack, noting that existing models often only predict outcomes in mice. While industry leaders like Anthropic’s Dario Amodei and OpenAI’s Sam Altman have touted AI’s potential to cure cancer, actual results remain limited, with few AI-designed drugs advancing to late-stage trials. Vivodyne, spun out of the University of Pennsylvania, recently opened a large facility near San Francisco. The company reports high predictive accuracy for its liver, airway, and bone marrow tissues compared to human trials. Having raised nearly $80 million, Vivodyne aims to provide pharmaceutical partners with reliable human-relevant data before expensive clinical trials, potentially reducing the high failure rate of drugs that succeed in animal tests but fail in humans.
💡 Why It Matters
- · Vivodyne challenges the industry’s reliance on static cellular snapshots by providing dynamic, causal data from living human tissues.
- · This approach directly addresses the critical bottleneck where 90% of animal-effective drugs fail in human trials, offering a tangible path to validate AI predictions before costly clinical testing begins.
