Q13. Where does the data actually come from, and can we trust it's collected legally and structured well?

Every source goes through a vetting process before it's part of the corpus. That means checking more than just robots.txt: privacy policies, legal notices, and terms and conditions, to confirm the data is being collected and used the way each source allows. It's best-effort, and if you ever think something looks off, our data contact is listed on our website and we want to hear about it. Once a source clears that bar, it feeds into a corpus with obligations clearly defined rather than raw scraped text, which is also what makes it possible to plug into advanced applications inside your own systems.