MyTechDailyNews General Report: How Claude and six organizations Are Testing AI at Every Step of Drug Development
A general-reader view of a long, distributed problem
For people who follow technology as a story about real workflows, the interesting part of this Anthropic update is not the promise of a quick cure. It is the attempt to put Claude inside the slow, administrative, and scientific steps that can delay medicines for years. MyTechDailyNews treats this as a product-development question: what changes when an AI assistant is asked to move across research, design, operations, documentation, and planning?
The update says: Over the past several months, Anthropic’s life sciences team entered deep partnerships with six organizations that are developing new medicines. The named groups are BigHat Biosciences, Bristol Myers Squibb, Divergence Labs, Enveda, Generate Biomedicines, and The Column Group. Anthropic has also started its own programs, including rare childhood forms of muscular dystrophy, where Claude handles much of the scientific and operational work. It plans to scale that work substantially.
Drug development is often reduced to one visible challenge: proposing a new molecule. The announcement describes a wider set of tasks, including finding a biological target and choosing the drug type that can engage it, designing molecules with useful properties, making reliable manufacture possible, preparing clinical trial filings, and planning or running trials.
Anthropic’s strategy is to spread AI use across that chain instead of looking for one decisive breakthrough. The update frames the work as a series of distributed bottlenecks. It also describes the purpose in three ways: the partnerships reveal model strengths and gaps across many tasks, they create case studies for the wider life sciences community, and—if the programs keep improving—they may help produce effective new medicines for food allergies, rare childhood diseases, and several forms of cancer. Anthropic says wider adoption by many more biotechnology firms is necessary if Claude is to significantly increase the rate of new cures.
Project-by-project updates
Anthropic says technical reports and preprints describe the thought process, methods, and early results for several partner programs.
Generate Biomedicines: editing a therapy’s instructions
Generate Biomedicines is working to extend CAR T-cell therapy beyond blood cell cancers, including solid tumors. The team shortened one part of the therapy’s genetic instructions so it could leave room for other instructions aimed at different cancer types. Six Generate scientists used Claude Code to propose new designs for that part. One design was about 70% shorter than the standard version, and in lab tests it performed better: immune cells made roughly 10 times as many receptors, the proteins that spot cancer cells, killed tumor cells more efficiently, and stayed active longer.
Divergence Labs: building a large antibody search
Divergence Labs is testing a different idea for food allergies caused by IgE. Conventional treatments only neutralize IgE moving through blood; the team wants molecules that remove IgE from immune cells. A stronger design would need two arms: one that binds tightly to IgE, and another that helps prise IgE from the immune cell. Claude turned the company goals into a detailed plan and then produced, largely autonomously, roughly 35,000 antibody designs in about a month for narrowing and testing. Early tests identified hits for both roles: three designs that could bind to IgE, and 20 that could dislodge it. Three of those 20 appeared to remove IgE better than the standard medication. The next step is to combine the two arms into one drug.
BigHat Biosciences: connecting models to an automated lab
BigHat Biosciences pairs AI with an automated lab to improve candidate medicines called leads. Leads need to attach to targets tightly, stay stable during production and storage, and avoid sticking together. Claude was asked to support these three properties in two leads, both antibody fragments that could later become parts of drugs: one aimed at cancer cells, one aimed at making a drug persist longer in the body. Connected to Reccy, BigHat’s lab data system, Claude suggested molecule changes, and the lab made and tested about 1,010 variants in about 10 weeks. Midway, the team doubted standard heat-stability and clumping measures. Claude and an Anthropic researcher reanalyzed raw data, proposed alternate metrics and a protocol change; BigHat scientists refined and ran it, and the main predictions held. The best versions bound more than 10x more tightly, the cancer-targeting lead stayed in solution 18°C higher, and the longer-lasting lead clumped more than 100x more slowly.
Enveda: using AI for quality documents and filings
Clinical trials require batch-quality records and regulator filings. At Enveda, manufacturing chemists spent up to 75% of their time on paperwork. Enveda linked all manufacturing documents to Claude and asked it to flag potential problems. In tests across 1,061 pages of documentation, Claude caught all 11 issues that human reviewers had spotted—and flagged seven they’d missed. The review time fell by 89%, from over 25 hours to less than three. Separately, Claude wrote the first drafts of two important regulatory documents. One took four hours versus about three months by hand; human review of the second found one minor error.
The Column Group: TCG, Pleiades, and riskier drug ideas
Biotech investors often prefer validated ideas because discovery is slow, costly, and failure-prone, even when riskier concepts could help more patients. Anthropic has worked with The Column Group, called TCG here, to test whether Claude can lower the time and cost enough for investors to fund more ambitious ideas. TCG is establishing Pleiades, a company that will use Claude to run more—and more complex—drug programs than a small team usually handles. Pleiades is currently developing tools and workflows for Claude. Its very first Claude skill helps assess whether drug targets are worth pursuing, and it has been evaluated on 45 target-disease pairs.
Bristol Myers Squibb: planning and probability checks
Anthropic recently started work with Bristol Myers Squibb to find bottlenecks in clinical trial planning. BMS is applying Claude to early-stage trial plans and success-likelihood estimates. The teams plan to expand this work in 2027 to see whether it can raise the probability of success in early development.
We use Claude at the clinical trial stage to analyze proprietary data, create a clinical study plan, and weigh the study’s chance of success. From a single prompt, Claude generated a robust initial analysis that helped inform our assessment of different development options, which would otherwise take days to complete. We did all of this while ensuring the security of our highly sensitive data. We look forward to expanding this work across our pipeline to support portfolio-level decisions.
Michael Pourdehnad, MD, Senior Vice President, Head of Early Clinical Development, Hematology, Oncology and Cell Therapy.
Mithril Medicine: a small company inside a larger lab
Anthropic acquired Mithril Medicine, a one-person company developing treatments for two rare forms of muscular dystrophy. Founder Cody Tranbarger used Claude to organize contracts, logistics, and experiments with 11 external labs and delegated many operational tasks. Mithril took a possible treatment for one muscular dystrophy form from concept to promising mouse results in less than 18 months. At Anthropic, Cody continues the research and plans several new programs. If candidates hold up, Anthropic expects a drug ready for human clinical trials as soon as late 2028.
Checklist for general readers
- Who is involved: six organizations named in the update, plus Anthropic’s own programs.
- What Claude is touching: target discovery, molecule design, optimization, manufacturing paperwork, filings, company creation, and clinical trial planning.
- What has been observed so far: shorter genetic instruction segments, autonomous antibody design batches, better lead properties, faster document review, and faster regulatory drafts.
- What remains ahead: most work is preclinical, and the next test is how programs behave as they move toward human trials.
Common misconceptions
| Misconception | Why the update does not support it |
|---|---|
| AI has already approved a treatment. | The announcement says the research is early and mostly preclinical; for the most part, drugs are not yet being tested in humans. |
| One AI model can solve drug development. | It describes distributed bottlenecks across science, manufacturing, paperwork, and trial planning. |
| Only laboratories matter. | The case studies include document checks, regulatory drafts, company creation, and trial assessment. |
Frequently asked questions
Which organizations are part of the early program?
Anthropic says it entered partnerships with six organizations: BigHat Biosciences, Bristol Myers Squibb, Divergence Labs, Enveda, Generate Biomedicines, and The Column Group.
Is the work only about designing molecules?
No. The update lists target identification, molecule design, manufacturing optimization, filings, trial design, and trial running as separate areas where AI can support the process.
Can more researchers use these tools?
The update says Anthropic has rolled out Claude Science, an AI workbench for researchers, and is providing Claude credits for life scientists. It has also created the Life Sciences Verification Program so trusted organizations can use its most powerful models with more permissive versions of safeguards needed to prevent misuse but that can impede legitimate research. Anthropic recently launched an open call for guest investigators in life sciences.
Action suggestions
- Read health-AI announcements as workflow evidence: ask which task the model handles, what was measured, and what stage of development is affected.
- Teams comparing AI tools should look beyond model chat, including document review, filings, company operations, and trial planning, because bottlenecks are distributed.
- For readers following medicine, track whether preclinical work becomes clinical trial readiness; the update treats late 2028 as one possible timeline for a rare muscular dystrophy program.
What MyTechDailyNews will watch next
The story is not a single product launch; it is a series of early deployments. The clearest takeaway is that Anthropic is using real drug programs to discover where Claude helps, where it needs better tools, and how administrative drag can be reduced. The next phase depends on two things: whether partner and internal programs advance toward clinical-stage decisions, and whether Claude Science, Claude credits, the Life Sciences Verification Program, and the open call for guest investigators let many more teams learn from these early cases. For general tech readers, that is the practical significance: AI is being tested not only as a source of ideas, but as a layer inside the slow machinery of bringing a possible medicine forward.