Anthropic says about 950 Claude agents used 210 million tokens over 21 hours to survey more than 200,000 reverse transcriptases, identify 3,500 candidate systems and narrow them to 20 reports. Human scientists tested the strongest candidate in a BSL-1/2 lab. The resulting array-associated reverse transcriptase system combines an RT, a partner gene and a repeated DNA array; early experiments show the array is expressed as short RNAs. Its biological function remains unknown and further experiments are ongoing.
Claude agents surface a new reverse-transcriptase system for lab testing
2026-09-23Frontier watchWATCH
Anthropic says about 950 Claude agents used 210 million tokens over 21 hours to survey more than 200,000 reverse transcriptases, identify 3,500 candidate systems and narrow them to 20 reports. Human scientists tested the strongest candidate in a BSL-1/2 lab. The resulting array-associated reverse transcriptase system combines an RT, a partner gene and a repeated DNA array; early experiments show the array is expressed as short RNAs. Its biological function remains unknown and further experiments are ongoing.
Imported from the existing research archive. Migration preserves the recorded judgment and is not a fresh verification.
What this means for you
For research-agent builders, this is a rare end-to-end example where a large agent fleet generated a hypothesis that survived human review and reached wet-lab testing. It does not show autonomous scientific validation: humans designed and performed the experiments, the study is vendor-run, and the function is unresolved. Treat it as a workflow pattern to watch, not proof of a CRISPR-scale discovery.
FRONTIER WATCH
Public artifacts and validated results are different.
The recorded fact is the public work and its artifacts. Author-reported outcomes do not establish independent reproduction. Next checks include usable code, task conditions and third-party validation.
Evidence & related material
Technical details & scores
| Agent Run | about 950 agents, 210M tokens, 21 hours |
|---|---|
| Search Space | >200,000 reverse transcriptases; 3,500 candidate systems; 20 final reports |
| Candidate | array-associated reverse transcriptase (ART): RT, partner gene and repeated DNA array |
| Validation | human-run BSL-1/2 lab work; array expressed as short RNAs |
| Limit | biological function unknown; vendor-run preprint and ongoing experiments |
Technical impact 4
Engineering value 3
Adoption signal 2
Maturity 2
Verification cost 5
Risk 4