Named CRISPR and genome-mining researchers are pushing back on Anthropic's Claude ART announcement in the same news cycle. Spotting a familiar DNA pattern in public databases is not the same as proving a new editing tool. Anthropic itself still says ART's function is unknown.
THE DECODER reported the skepticism on Sept. 24, 2026, framing the company claim as routine genome mining rather than a demonstrated breakthrough. Decoder paraphrases Lucas Harrington, a Doudna-lab Ph.D. and Mammoth Biosciences co-founder, who posted on X that genome-mining methods have been used for decades and that similar systems have been known since 2008. Harrington's locked point, via Decoder, is that the hard part is showing what a system does, which Anthropic has not shown, and that presenting early results as a major discovery is not helpful.
Anthropic's Sept. 23 claim credits Claude agents with flagging ART, an array-associated reverse transcriptase system with CRISPR-like DNA repeats mainly in phage genomes. The company says the underlying reverse transcriptase was already known. The claimed novelty is the linked repeat array and partner gene. Function remains unknown on that company account.
Trending Topics, also Sept. 24, quotes Washington University microbiologist Kevin Blake saying nothing indicates ART is a rival to CRISPR-the-technology. The same piece notes the reverse transcriptase itself was previously known, with Claude credited for linking it to the overlooked array.
What remains open is wet-lab characterization of what ART actually does, independent replication beyond Anthropic's company-hosted preprint, and whether further researcher pushback hardens into a consensus frame. Until function is shown, the story is a same-cycle skepticism loop around a genome-mining claim, not proof of a next CRISPR platform.