Claude is moving into real drug programs, and the early numbers are hard to ignore

Manufacturing chemists at one biotech firm were spending up to 75% of their working hours on paperwork. That figure says a lot about where drug development actually stalls, and it is not where most AI headlines point. Anthropic's latest update suggests the slowdowns sit all along the pipeline, and an AI model can reach many of them.

What's the news?

Anthropic's life sciences team has spent several months working with six organizations: BigHat Biosciences, Bristol Myers Squibb, Divergence Labs, Enveda, Generate Biomedicines and The Column Group. It has also started drug discovery programs of its own, including work on rare childhood forms of muscular dystrophy. The company says it plans to scale this up substantially.

The results so far are specific. At Generate Biomedicines, six scientists used Claude Code to design a shorter part of a CAR T-cell therapy, aimed at extending it to solid tumors. One design was about 70% shorter than the standard version. In lab tests, immune cells made roughly 10 times as many receptors, killed tumor cells more efficiently and stayed active longer.

Divergence Labs, which is developing a food allergy treatment that pulls IgE off immune cells, had Claude generate about 35,000 antibody designs in roughly a month. Early tests found three designs that bind IgE and 20 that dislodge it. Three of those appeared to outperform the standard medication.

At BigHat, Claude worked with the company's lab data system to propose changes to two antibody fragments. About 1,010 variants were made and tested in about 10 weeks. The best versions bound more than 10 times more tightly, and one clumped more than 100 times more slowly.

Enveda's results were the most striking on time. Claude caught all 11 issues human reviewers had found in 1,061 pages of manufacturing records, and flagged seven they had missed. Review time fell 89%, from over 25 hours to under three. One regulatory draft took four hours instead of about three months.

Why does it matter?

Most public discussion of AI in medicine centers on proposing new molecules. This work argues that the bigger gain may come from the unglamorous tasks: target research, lead optimization, batch records, regulatory filings and trial design. Because the bottlenecks are spread out, speeding up one step is rarely enough.

There is also a financial angle. The Column Group is setting up a company called Pleiades to run many more, and more complex, drug programs with a small team. If drug discovery gets cheaper and faster, investors may back bolder ideas instead of retreating to validated ones.

Caution is still warranted. These are early results, mostly from lab tests and internal reviews. None is a approved medicine, and clinical trials remain the real test.

The context

For the GCC, the timing is relevant. Saudi Arabia's Vision 2030 and the UAE's life sciences and genomics programs both depend on building local biotech and clinical research capacity. Regulatory documentation and manufacturing compliance are areas where smaller regional firms often lack headcount. Tools that cut a three-month drafting job to hours could change what a small team can attempt.

So the practical question for regional health leaders is not whether AI will touch drug development. It is how quickly Gulf hospitals, regulators and biotech start-ups can build the data systems and review processes that let them use it safely. Anthropic says it wants to work with many more biotechnology firms. Regional companies should expect to be part of that conversation.