Anthropic's AI Model Claude Discovers Novel Enzyme System Resembling CRISPR

AI company Anthropic announced on Monday that its large language model, Claude, has discovered a novel enzyme system with characteristics similar to the groundbreaking CRISPR gene-editing technology. The finding emerged from the early work of a new life sciences research group and laboratory the company established to explore how AI can accelerate fundamental biological research.

AI Accelerates Biological Discovery

The discovery highlights the potential for AI models to dramatically speed up scientific analysis that would typically take human experts weeks or months. Anthropic's research team, formed in spring 2026, tasked Claude with searching through a vast database of DNA sequences to identify new systems.

The newly identified system is called array-associated reverse transcriptases (ART). According to a pre-print released by Anthropic, the system consists of a reverse transcriptase enzyme, a partner gene, and an array of DNA repeat sequences, a layout that resembles CRISPR arrays. Initial experiments have confirmed that the ART array produces distinct short RNAs.

How Claude Found the Enzyme System

While human scientists provided initial direction and conducted subsequent laboratory experiments, Claude performed the heavy lifting of database analysis. The process involved:

Ad

• Using 950 AI agents working over 21 hours.
• Processing 210 million tokens of genetic data.
• Identifying over 200,000 reverse transcriptases and narrowing them down to 3,500 candidate systems.
• Further refining the list to 20 candidates for detailed analysis by scientists.

While the underlying reverse transcriptase enzyme had been identified in previous studies, Anthropic stated that Claude was the first to recognize the system's defining features, including the associated array of non-coding DNA sequences and an additional accessory protein.

Expert Validation and Future Research

The discovery has been reviewed by a leading expert in the field. Feng Zhang, a pioneer of CRISPR genome editing and a professor at MIT and the Broad Institute, called the finding an "exciting example of how AI agents can contribute to biological discovery" in a review of the pre-print.

Zhang added that the identification of these systems is "genuinely intriguing and merits further investigation." Anthropic confirmed that its work to understand the primary function of ARTs is ongoing. The announcement underscores the increasing convergence of artificial intelligence and life sciences, a trend investors are watching closely for its potential to disrupt drug discovery and biotechnology.