Anthropic's high-profile claim of a biology breakthrough powered by its Claude AI is drawing scrutiny from at least one scientist who says the enzymes at the center of the announcement were already under active investigation long before the company's researchers got involved. The dispute highlights a growing tension between the pace of AI-driven science communication and the slower, less glamorous work happening in traditional research labs.
What Anthropic Claimed
Earlier this year, Anthropic announced that its internal biology lab had used Claude to identify and analyze a novel class of enzymes, framing the find as a significant step forward in AI-assisted biological research. The announcement attracted considerable media coverage and was cited as evidence that large language models could contribute meaningfully to scientific discovery. Coverage from outlets including New Scientist pushed back somewhat on the framing, with reporters noting the enzymes might be better described as another gene-editing tool rather than an unprecedented find. That reporting aligned with concerns now being raised directly by a working scientist.
Key Facts
- A scientist says he has studied the same enzymes for approximately four years.
- Anthropic's biology lab announced the find as a major AI-assisted discovery.
- The dispute centers on whether the enzymes were previously unknown or already under investigation.
- Anthropic has not yet issued a detailed public response to the scientist's claim.
- The episode follows earlier coverage questioning the novelty of the lab's initial findings.
The researcher in question has stated publicly that his lab was already deep into studying these enzymes well before Anthropic made its announcement. He did not accuse anyone of deliberate misrepresentation but said the framing of the discovery as a breakthrough was, at minimum, misleading. His comments have spread quickly through science and AI circles, prompting broader questions about how AI companies communicate research findings to the public. Stories like this one are part of a wider pattern covered across the latest Claude AI news, where announcements from AI labs meet resistance from domain experts.
"We have been studying these enzymes for four years. To see them described as a new discovery is frustrating for everyone who has put in that time."Scientist responding to Anthropic's biology announcement, via social media
A Pattern Worth Watching
This is not the first time Anthropic's biology research arm has faced questions about the significance of its reported findings. When the lab reported a major early discovery, observers noted the announcement leaned heavily on superlatives while the underlying science was still being assessed by outside researchers. Critics argue that AI companies have an incentive to frame any positive result as a breakthrough because it supports fundraising narratives and public perception of their technology's capabilities.
Anthropic's biology work is conducted through a dedicated internal lab that pairs Claude's analytical capabilities with human researchers. The company has argued that AI can compress research timelines and surface patterns that human scientists might miss. That argument has merit in certain contexts, but it does not address the more specific allegation here, which is that the enzyme work was not novel in the way it was presented. Questions about novelty and attribution are not unique to AI research, but they become sharper when a technology company is making the claim rather than a peer-reviewed journal. The New Scientist coverage characterizing the find as another gene-editing tool suggested skepticism was building even before this latest challenge emerged.
For now, Anthropic has not responded in detail to the scientist's account. The company's communications team has pointed to its published research as the authoritative record. Whether that will satisfy critics is unclear. What is clear is that the episode adds a new layer of complication to how Anthropic's biology ambitions are perceived, both within the scientific community and among the broader public following AI development.
“When AI labs announce scientific breakthroughs, organisations must immediately ask who verified the claim and over what timeline. If a human researcher spent four years on identical work, that context changes how you should trust, adopt, and build strategy around AI-generated discoveries.”
Leon Tindemans, AI expert and entrepreneur specialising in Claude, Copilot and ChatGPT. Learn more with AI literacy training by TTM Communicatie.
