Kevin Weil, who previously led science efforts at OpenAI, has launched a new company that uses AI-trained robots to do the hands-on chemistry behind new medicines. The startup, Warp Sciences, has raised $215 million at a valuation of $865 million, TechCrunch reports. Weil announced the company and its funding on X on 22 October.
What the company does
Finding a new drug involves a lot of slow, repetitive lab work. Scientists identify chemical compounds that might work, then someone has to actually make them and test them in what is known as a “wet lab”, a lab where real chemicals and liquids are handled rather than computer models.
Warp Sciences wants robots to do that making and testing. Researchers point the robots at promising compounds, and the machines synthesise and test them. Because robots do not need breaks, the idea is that this work can run around the clock and speed up drug discovery.
How the robots learn
The robots are trained by watching people. According to TechCrunch, human chemists wear cameras and gloves fitted with sensors while they carry out lab tasks. That recorded data is then used to teach the robots to copy the same steps. It is a similar approach to how some companies are training robots for warehouses and homes: capture skilled human movement, then turn it into training data.
Bringing lab work home
There is also a supply-chain angle. Drug and biotech companies have long outsourced this labour-heavy chemistry to contract labs, including China-based WuXi AppTec. TechCrunch says Weil wants to bring that work back to the US and reduce the risk that political tensions disrupt drug research.
Who is backing it
The round was led by venture firm Lightspeed, with IVP, Mayfield, Lux Capital, Premji Invest and Menlo Ventures also taking part. The size of the round, for a company that has only just been announced, shows how keen investors are on AI that is applied to science and medicine rather than to chatbots.
It is worth keeping expectations in check. The company is brand new, and TechCrunch’s report is based on Weil’s own announcement, without published results showing how well the robots perform yet.
What it means for you
New medicines typically take many years to develop, and the early lab stage is one of the bottlenecks. If robot labs like this work as promised, they could help researchers test more ideas faster, which over time could mean more treatments reaching trials. For lab workers and chemists, it points to a future where some routine bench work is automated, and where skilled people may increasingly be needed to train, supervise and check the machines rather than do every step by hand.