The underlying TechCrunch AI report highlights a core tension in Anthropic's strategy. While the company publicly advocates for caution and safety testing, it is simultaneously pushing aggressively into high-stakes domains like wet-lab biology, an area CEO Dario Amodei has flagged for bioterrorism risks. The framing of this discovery—'mostly' by Claude—serves as a powerful marketing tool for the model's capabilities, but the actual lab work remains human-operated for now.
The key question is whether this represents a cautious, incremental step or a prelude to the fully automated labs Amodei has not ruled out, a move that would dramatically scale both potential and peril.