A Revolutionary Approach to Cancer Research

A biotech startup called Vivodyne is challenging the conventional wisdom that AI will soon cure cancer. The company’s CEO, Andrei Georgescu, argues that the AI drug-discovery industry is hindered by a critical data problem. Current AI models are trained on data from animal testing, single cells, or proteins, but not on living tissue. This limitation prevents AI models from accurately predicting the efficacy of potential cancer treatments.

Source: techcrunch.com

Vivodyne’s innovative solution is called HIVE, a modular robotic lab that can grow 20 different types of human tissue. The HIVE system can autonomously dose and monitor the tissue, generating the kind of causal biological data that today’s AI models are missing. This data is essential for developing effective cancer treatments.

Georgescu emphasizes that the absence of human testing data is a significant limitation for AI models. ‘They’re going to cure cancer in mice,’ he says, highlighting the gap between AI predictions and real-world outcomes. Even prominent figures in the AI community, such as Anthropic CEO Dario Amodei, have expressed skepticism about the prospect of AI curing cancer.

The Challenge of Data Scaling

The pharmaceutical industry is facing a similar challenge. A staggering 90% of drugs that are effective in animal testing fail to receive regulatory approval for humans. Vivodyne’s HIVE system aims to address this issue by providing a more accurate representation of human biology. The company claims that its tissues closely match the behavior of real human organs, with liver cells achieving 94% predictive accuracy and airway tissue matching the behavior of real human tissue 96% of the time.

Georgescu believes that Vivodyne’s approach can accelerate the path of drug candidates by providing a better understanding of what will work before going through the expense of a clinical trial. The company has already achieved twice the throughput of all the animal trials being held in the US and is working with multiple major pharma companies to solve this problem.

A Larger Vision for Human Biology

Georgescu’s vision extends beyond curing cancer. He sees Vivodyne’s autonomous biology labs as key to generating the kind of causal data that can be used to train new models on human biology. He points to studies that find no clear data scaling laws when training generative AI models on existing cellular data. ‘All the training is done on static snapshots of these cells, and the models are not conditioned at all by the how a cell got to that state,’ Georgescu explains.

Vivodyne’s HIVE machines are tracking hundreds of thousands of ongoing experiments where diseased tissue is exposed to some stimulus. Georgescu expects this to provide the kind of reinforcement learning that will produce AI models that understand human biology enough to make more meaningful progress in healthcare.

The Future of Cancer Research

Georgescu believes that Vivodyne’s approach will be crucial not just for today’s medicine challenges but also for a future where complex diseases require drugs that target multiple pathways. ‘If we want combination therapies, the space that has to be searched explodes—it can’t be an experimental approach,’ he says. Establishing causality in human biology is the basis of all this, and Vivodyne’s HIVE system is a significant step in that direction.

In conclusion, Vivodyne’s innovative approach to cancer research is a promising development in the field. By providing a more accurate representation of human biology, the company’s HIVE system has the potential to accelerate the path of drug candidates and make more meaningful progress in healthcare.