Automating a single stage leaves the rest of the chain untouched. If a machine runs the experiments but a person still chooses the target and interprets the results, the lab as a whole is not autonomous. The automated node may speed up its own step while the surrounding steps remain human-dependent, and the handoffs between them still require judgment. This is why automating a task and running a lab are different achievements, and why the honest answer to whether AI can run a lab on its own is a matter of degree.
Can AI Run a Drug Discovery Lab on Its Own?
What It Would Mean for AI to Run a Lab
Why One Automated Step Isn't Enough
3 / 3
Now suppose you automate just one of those stages. Say a robot runs the experiments perfectly, every time. Does that make the lab autonomous?
0:00 / 0:00