A young startup is spending hundreds of millions to prove that artificial intelligence can salvage stalled drug pipelines.

Pathos AI is not just building software. It is buying its way directly into the clinic.

The company is paying $125 million upfront to license a Phase 3-ready cancer drug from Alphamab. Simultaneously, it is taking over the early-stage clinical development of an AstraZeneca breast cancer degrader.

This is a massive financial gamble for a company founded in 2022.

The Trial Design Gamble

Most AI drug discovery companies focus on finding new molecules. Pathos is taking a different route. It is using its multimodal AI platform to redesign clinical trials for existing drugs.

By taking over AstraZeneca’s asset, Pathos is testing its platform on a global scale. The company’s leadership, heavily drawn from Tempus, believes data-driven patient selection can rescue drugs that might otherwise fail broad trials.

The goal is to identify which specific patient subgroups will actually respond to these therapies. If the AI is right, Pathos resurrects a drug. If the AI is wrong, the company burns through its capital on expensive clinical trials.

High Stakes Funding

To fund this strategy, Pathos is already hunting for a $300 million Series E round. This comes just a year after a $365 million Series D.

This rapid burn rate highlights the reality of modern biotech. AI does not make clinical trials cheap. It only attempts to make them smarter.

Whether Pathos can successfully pick winners out of other companies’ pipelines remains unproven. But the industry is watching. If this model succeeds, it could change how pharma handles stalled drug candidates.