Anthropic's Claimed AI Lab Discovery Draws Challenge From Biologists

According to Anthropic, the system used 950 agents and after 21 hours flagged a repeating pattern surrounding a known enzyme. The pattern had not been catalogued before, the company said. It was not a brand-new DNA sequence. Anthropic described the pattern as “reminiscent” of what led to the gene-editing technology CRISPR, which the company said “has already transformed science and medicine.” That framing helped make the result sound like a significant discovery.

Some biologists disagreed. Lucas Harrington, the biologist who wrote a viral post criticizing Anthropic’s announcement, said that “finding a weird cluster of genes and repeats is often the easy part. The hard part, and where the real discoveries come from, is figuring out what the system actually does.” The post was subsequently endorsed by the chair and CEO of the drugmaker Eli Lilly. In this view, the agents helped with laboratory grunt work, but did not make a discovery.

The dispute grew more complicated when Mario Rodríguez Mestre, a biologist at the University of Copenhagen, said over the weekend that his team had already discovered the particular pattern, the New York Times reported. Mestre, who regularly chatted with Claude in his work, wondered whether Anthropic’s team had learned from his conversations. Anthropic denies this, but Mestre says he is stopping all use of Claude anyway.

Part of the problem, according to the article, is that AI companies are not presenting their systems simply as tools scientists can use, like microscopes or supercomputers. They are insisting that the AI systems are making discoveries themselves. To some, that approach is incompatible with how science actually works, with new knowledge more typically emerging from collaboration and an ever-growing arsenal of tools. It also makes people more skeptical of genuine progress when it happens.

Whittling 200,000 candidates down to a few worth exploring is no small feat and is legitimate scientific work, the article said. The fact that a general-purpose chatbot could do that work is notable, even if humans helped steer it and ultimately ran the experiments. But once the standard becomes whether Claude itself made a discovery, all of that becomes evidence for one side or the other in a debate with only two answers: breakthrough or bust.

A similar dynamic followed OpenAI. Earlier this month, OpenAI said its own team agents had cracked a million-dollar problem in mathematics. A couple of weeks later, AI skeptics were sharing an article asking whether it was the math problem that really mattered, according to MIT Technology Review. The piece did not argue that OpenAI’s solution was wrong. Instead, it argued that the particular result may not be the one mathematicians care most about. A mathematician also accused the models of possibly using some of his work without credit, leaving people to think either OpenAI cheated or the solution was not important anyway, or both.

Harrington closed his post with a suggestion: AI companies should “set the bar high now, so that when an AI actually discovers a fundamentally new biological mechanism, everyone appreciates how big a deal it is.” But as OpenAI’s Sam Altman and Anthropic’s Dario Amodei race to one-up each other, raising the bar for scientific breakthroughs by AI might be the last thing on their minds.

Editor's Summary

Anthropic says its Claude-powered biology lab found a previously uncatalogued genetic pattern, but biologists dispute whether that amounts to a scientific discovery, and a Copenhagen researcher says his team found the pattern first. The dispute highlights a broader debate over AI companies presenting models as discoverers rather than tools, as similar skepticism followed OpenAI's claimed math result. The debate centers on whether AI systems are tools that assist scientists or independent discoverers, and on what evidence should count as a genuine breakthrough.