I was asked to scrape some prediction website millions of times every day. Instead of doing that, I reverse engineered their prediction curves from a few hundre

I was asked to scrape some prediction website millions of times every day. Instead of doing that, I reverse engineered their prediction curves from a few hundred data points and served the models myself. Not sure which one is more ethical. Scraping at scale or stealing the models using mathematics. But I know second one is cooler.