You didn’t finish my comment. Read the last sentence.

Anti-fingerprinting tech just produces a different fingerprint. Google knows e.g. when things are scrambled but certain other things stay the same.

Untrue, you can modify it enough to avoid giving it more entropy. Possible approaches include: - Spoofing browsers down to the TCP stack - Plausibily random values - Every possible bit scrambled on each request

You can see a similar thought-process behind Tor bridges so it is tried-and-tested. Noted that it is a much more difficult feat to accomplish in a full blown browser rather than network layer.