If fake data is a concern, then you can use a headless browser to scrape data. Headless browsers see content exactly the same way as the end user. There are no ways to detect headless browsers, because… well, they behave as regular Google Chrome or Firefox browsers. For a quick start, see https://pptr.dev/.
For the record, this is a lazy approach, and not one that I would implement. You should be very much able to distinguish fake from real data using the markup.