At this week's “The Briefing: AI for Science,” Anthropic unveiled its plans to develop its own drugs, aiming to leap from coding tools into healthcare. The move puts it alongside other tech giants like OpenAI and Google in the race towards medical breakthroughs. But will AI-driven drug discovery really revolutionise medicine, or is it just another buzzword?

Anthropic’s Eric Kauderer-Abrams highlighted neglected diseases as targets, but details are sparse. The company’s AI prowess could aid research by suggesting new compounds or improving known ones. However, the path to regulatory approval remains a steep challenge.

Experts say AI is already integral in drug discovery, from generating ideas through clinical trials and manufacturing. But actual drug development still requires extensive testing, skilled labour and years of effort. The journey from lab to pharmacy could take over a decade.

The AI boom in drugs might not live up to its hype. Professor Matthew Todd cautions that while AI can speed up research, it won’t replace the need for rigorous experimentation and clinical trials. Anthropic will need to invest heavily in experiments if it wants to deliver on its promises.