Hi Dotikers!

Big news from the AI world this week: Anthropic claims its model Claude has autonomously discovered a new enzyme system hidden in bacterial DNA, one that looks suspiciously like CRISPR, the gene editing tool that earned Emmanuelle Charpentier and Jennifer Doudna a Nobel Prize in 2020. The headline stat is impressive: Claude combed through a massive database of DNA sequences and flagged the pattern in just 21 hours. Dario Amodei, Anthropic's CEO, is already talking about AI curing most diseases within a decade.

Let's take a breath here. What Claude actually found is a "CRISPR-like" sequence with an unknown function. Scientists have long known that millions of unstudied bacterial species are full of these structures, and finding one is a far cry from building a therapy around it. Microbiologist Kevin Blake put it plainly: nothing suggests this is a rival to CRISPR as a technology. Spotting a promising pattern in nature and turning it into a Nobel-worthy tool are two very different sports, a bit like confusing finding a guitar in your attic with headlining a stadium tour.

That said, dismissing this entirely would be a mistake. Stanford's Stanley Qi highlighted the real breakthrough: an AI recognising an unusual biological pattern that humans struggled to detect, then pursuing it as a genuine research question. That capability, applied at scale, could compress years of exploratory biology into days. The discovery itself may turn out to be a footnote, but the method behind it is the story that matters.

So yes, the announcement is oversold, timed conveniently amid Anthropic's push on AI regulation and its rivalry with other frontier labs. But underneath the marketing gloss sits something genuinely promising: AI as a tireless research assistant, scanning the vast unexplored corners of biology. Keep the champagne on ice, but do keep watching this space.

Alex.