Kevin Weil’s new company, Warp Sciences, has raised $215 million to automate laboratory work for drug discovery, using robots to make and test compounds. The sizeable funding backs a plan to turn physical experiments into an AI training loop, not just another promise that software will speed up science.
Watch Desk analysis
What happened
TechCrunch reports that the company raised the money in a Series A led by Lightspeed, at an $865 million valuation. It plans to operate as a chemistry contract research organisation, supplying custom molecules to pharmaceutical and biotechnology partners.
The company intends to use data from round-the-clock automated experiments to train and improve its AI models. Its stated aim is to bring synthesis work often outsourced to contractors back in-house and reduce supply-chain exposure.
Why it matters
Drug discovery is not only a matter of predicting which molecule might work. Someone still has to make compounds and test them. Warp Sciences is betting that robots can take on more of that laboratory work while generating data that improves its models. If it works, the advantage could be a tighter loop between an AI system’s predictions and experiments in the lab.
The funding is a serious opening wager, but it is not evidence that the approach has already made drug development faster or produced successful medicines. The useful test will be whether the company can deliver reliable molecules to partners and show that its automated experiments improve the research process.
Our read
This is a concrete AI-and-robotics play in a field where the physical work matters as much as the model. The interesting bit is the proposed feedback loop: robots do the experiments, and those experiments help train the AI. That is a more testable proposition than “AI will transform drug discovery”, which has had a good run as a headline and a less tidy relationship with the lab bench.
What to watch
- Whether Warp Sciences begins supplying custom molecules to pharmaceutical and biotechnology partners.
- How much of its laboratory work is automated, and what kinds of experiments it can run reliably.
- Whether the company reports evidence that its experiment data improves its models or speeds up research.
Discussion spark: Should AI drug-discovery companies be judged first on faster research, or on whether their work produces medicines that reach patients?
Sources and evidence
- Former OpenAI science lead launches Warp Science to automate wet labs with robots (22 October 2026, 18:38 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.