In the original Foundation books by Asimov, the conceit of "Psychohistory" was similar to the concept of machine learning for pricing: The future can be predicted _if people aren't aware of the prediction to change their behavior in relation to it_

This is similar to 'adverse selection' in real life & in Zillow's model. The article makes a nod to this, but seems to imply that if you train your model on that adverse selection, you can come out ahead after paying to learn about it.

To me that kind of misses the point. Adverse Selection isn't a static feature of the landscape you can identify and avoid, it is people understanding what you understand, adapting, and responding. Train your model with adversaries trying to beat it, then you'll maybe counter the specific first round strategies they use, and they'll learn new ones and beat your new model with their 2nd round strategies. It's a continuous game. Your requirement to gather a corpus of training data will keep you in the 2nd turn of a game where the wins are biased to whoever has the 1st move.