Claude agents have identified a previously uncharacterized genetic system in bacteriophages, the viruses that infect bacteria. Anthropic calls it array-associated reverse transcriptases (ART): a reverse-transcriptase gene next to a partner gene and a long array of evenly spaced, non-coding DNA repeats. The arrangement resembles a CRISPR array, and the repeats are transcribed into short RNAs. But no one has shown that ART edits DNA, cuts genetic material or can be programmed like a gene-editing system.
Anthropic’s own central caveat is blunt: “Although we don’t yet know its function.” The finding is therefore best understood as AI-assisted genome mining that produced a promising biological hypothesis, not a new CRISPR replacement or a ready-made medical technology.
As an Amazon Associate I earn from qualifying purchases.
What Claude actually found
ART combines three features that had not previously been connected as one system:
- A reverse transcriptase (RT): an enzyme that copies information from RNA into DNA.
- A neighboring partner gene: a nearby gene that may participate in the same biological process.
- A long repeat array: evenly spaced, non-coding DNA repeats with a layout that visually recalls a CRISPR array.
The viruses carrying these systems are bacteriophages, which infect bacteria rather than people. Anthropic says the RT itself appeared in earlier research; WIRED reports that a 2021 paper on jumbo phages had already identified that enzyme. The claimed advance is recognizing the surrounding repeat array and partner gene as a coherent system, not inventing an entirely unknown enzyme.
How the AI search was run
Anthropic says approximately 950 Claude agents worked for 21 hours and processed 210 million tokens. They gathered more than 200,000 reverse transcriptases, selected 3,500 candidate systems and narrowed those to 20 candidates for detailed, human-readable reports. Anthropic says an expert scientist might need weeks or months to perform comparable genome-mining analysis. Those scale figures are company-reported, not an independently audited benchmark.
The agents did not independently complete a biological discovery from prompt to proof. Human scientists supplied the initial prompt, reviewed the candidates, designed and performed the laboratory experiments and interpreted the results.
What has been observed in the laboratory
Anthropic’s initial experiments found that the ART repeat array is expressed as distinct short RNAs. That establishes that the repeats are transcribed. It does not establish what the RNAs do, whether they guide the reverse transcriptase, or whether they direct any reaction on DNA or RNA.
Free tools Windows power users keep installed
One-click scans. No signup required.
Rank #2
WIRED reported that the technical report had not been peer-reviewed and that it remains unknown whether ART can be turned into a gene-editing tool or would be useful if it can. More experiments are needed to establish biochemical activity, target recognition and reproducibility.
Why the CRISPR comparison is limited
CRISPR systems commonly use repeated DNA and intervening spacer sequences to encode guide information for programmable molecular machinery. ART’s repeat architecture is analogous in appearance, and its array produces short RNAs, but resemblance is not proof of CRISPR-like cutting, copying or pasting.
Anthropic says only a handful of known systems share the full combination of features and that those systems are programmable. That makes ART an interesting candidate for investigation, not evidence that ART itself is programmable.
Rank #3
| Question | ART evidence so far | What is not established |
|---|---|---|
| Structure | Reverse transcriptase, neighboring partner gene and long repeat array | How the components interact molecularly |
| RNA production | Repeats are expressed as distinct short RNAs | Whether the RNAs guide or regulate an enzyme |
| DNA editing | No editing activity reported | Cutting, insertion, copying or replacement of DNA |
| Programmability | Suggested only by analogy with a few related systems | Designed targeting, specificity and useful control |
| Novelty | System-level association appears newly characterized | The RT itself being previously unknown |
What scientists are saying
The reactions reflect two different judgments: enthusiasm for the search method and caution about the biology.
Feng Zhang, a professor at MIT and the Broad Institute, called the finding “an exciting example of how AI agents can contribute to biological discovery” and said the RNA-repeat arrays associated with reverse transcriptases merit further investigation.
Le Cong of Stanford told WIRED, “The experiments are still in the queue. The PR is already live.” Seth Shipman of the Gladstone Institutes said, “The novel thing is how they found it, not what it is.” Fyodor Urnov of UC Berkeley and the Innovative Genomics Institute said he sincerely complimented Anthropic for telling the world about the discovery.
Rank #4
What human researchers still have to determine
To establish a function, researchers will need to test the system’s components together and separately, identify the RNA molecules’ roles and measure the enzyme’s activity under controlled conditions. They would also need to determine whether any activity is specific enough to target chosen sequences, whether it can be programmed reliably and whether independent laboratories can reproduce the results.
Anthropic says its Bay Area laboratory operates at biosafety levels 1 and 2 and does not handle pathogens that infect humans. That safety boundary matters: the work concerns bacteriophage biology, not experiments on human pathogens or treatments for patients.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWhat this means for gene editing
ART could eventually reveal a new way that phages organize RNA-guided or reverse-transcription activity. It might also prove to be a biological curiosity with no practical editing use. At present, there is no evidence that it can replace established gene-editing systems, deliver a therapy, or make DNA changes in cells.
Best Value
The strongest demonstrated result is methodological: a large group of AI agents helped search an enormous sequence space and surfaced a pattern for human scientists to investigate. The scientific result remains provisional until experiments define what the system does and independent researchers verify it.
What to watch next
- Biochemical tests showing whether the reverse transcriptase acts with the repeat-derived RNAs.
- Experiments identifying the neighboring gene’s role.
- Measurements of sequence specificity, error rates and any DNA or RNA products.
- A peer-reviewed technical report and independent replication.
- Evidence, if any, that the system can be programmed rather than merely observed.
Until those results exist, calling ART “a new CRISPR” overstates the evidence. Claude found a distinctive genetic pattern and a plausible system around it; Anthropic’s scientists still have to discover what that system actually does.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
