Anthropic unveils Claude Science, an AI model built to speed up scientific research and drug discovery

US startup Anthropic has unveiled Claude Science, a new specialized model designed to help scientists find ways to fight diseases and develop drugs faster.

Why biotech and pharma are a priority for AI

Biotech and pharma have long been seen as one of the most promising fields for AI. Under the right conditions, AI can significantly speed up scientific discovery, from uncovering disease mechanisms to designing new drug molecules. Developing a new drug traditionally takes years and means testing a huge number of hypotheses, most of which get ruled out only in the late, most expensive stages of research.

That is why major AI labs are increasingly releasing models tailored to specific scientific and research tasks rather than general-purpose chatbots. For companies like Anthropic, specialized science models serve two purposes: they prove AI's practical value beyond writing text and code, and they open a market with deep pockets. Pharmaceutical companies and research institutes are willing to pay for tools that actually shorten development timelines.

What Anthropic offers scientists

Recognizing biotech's potential as an application area for AI, Anthropic has offered domain specialists Claude Science, a version of its technology focused directly on scientific work. The company intends the model to help researchers make progress in studying diseases and creating drugs by taking on part of the intellectual workload that used to demand months of manual work with scientific literature, experimental data, and hypotheses.

• Developer: US startup Anthropic
• Model name: Claude Science
• Target audience: scientists and domain specialists in biotech and pharma
• Stated goal: speed up the fight against diseases and drug development

Who else is competing for scientists?

Anthropic is not alone in entering the market for scientific AI tools. In recent years, leading AI labs have competed for the research community's trust, offering tools for analyzing scientific literature, working with molecular data, and automating routine experimental steps. For Anthropic, whose reputation rests largely on model reliability and safety, a move into biotech and pharma makes particular sense: mistakes in this field cost more than in most other AI applications, so a focus on accuracy becomes a competitive advantage.

Scientists traditionally spend much of their working time not on experiments themselves but on the routine intellectual work around them: searching for relevant papers, reconciling conflicting results from different labs, and formulating and discarding hypotheses before running expensive experiments. These are exactly the stages Anthropic expects Claude Science to ease, freeing up researchers' time for the more creative side of science.

What the company hasn't disclosed yet

Anthropic has not shared details about Claude Science's specific technical capabilities, the timing of a full release, or access terms for scientific organizations. Available information is limited to the fact of the announcement and the model's general purpose. That is typical for the early stage of bringing specialized science products to market: labs first announce a direction, then gradually reveal detailed capabilities, partnerships with research centers, and terms of use as the product matures and early feedback from scientists comes in.

What it means

The launch of Claude Science fits a broader trend: AI labs are competing not only for chatbot users but also for the trust of the scientific community, offering specialized tools for the most capital-intensive and socially important tasks, such as finding treatments for diseases that remain incurable. If models like this really do speed up the discovery cycle in biotech, the benefits will be felt far beyond the AI industry, in clinics and labs around the world.