A neutron spectrometer is like a blind geologist who taps on rock and guesses what's inside by the sound. It sends a beam of neutrons into a crystal and watches how they scatter. From this pattern, you can deduce how atomic magnets, spins, oscillate. But where exactly to dig and how long to listen used to be decided by a human.

Now, the TAS-AI algorithm has taken the helm. It works in two stages. First, like a curious novice: it pokes around almost at random, builds a rough map, and finds where the signal is strongest. Then it switches to expert mode: knowing the approximate location, it starts asking precise questions to determine which of two magnetic models is correct. In a test on a square lattice, it took fewer than ten measurements to overwhelmingly favor the right option.

The authors also noticed a pitfall: if the algorithm trusts its current guess too much, it stops looking for refutations. They added a "mandatory stress test" channel, and the system began spending less time on false leads. In a digital replica of the real instrument, the entire cycle works, and the code is openly available.