10 ms·
Why wouldn't a bot just use a proper browser (not headless), detect the "not a robot" box at the pixel level, and click on it using browser automation or some m
by _cs2017_ 8y ago
Why wouldn't a bot just use a proper browser (not headless), detect the "not a robot" box at the pixel level, and click on it using browser automation or some mouse movement script? At today's level of bot sophistication this seems almost trivial. Sure it might cost more in resources, but I doubt it's economically prohibitive when you're making at least a cent per fraudulent ad click?
- judge2020 8y agoI believe the Quora answer is putting a little too much faith in Google using "fair" factors when determining if a client is a robot. I'm sure it plays a part in determining "hijacked extension" activity vs human activity, but it's likely that the majority of the decision is how much recent activity your signed-in Google account has, whether or not you're signed in on Chrome (Firefox has a lot more stories of recaptcha challenges), and maybe even if you have Google WiFi or Google Home linked to your account. I wouldn't be surprised if they purely whitelist accounts that subscribe to Google Fiber or Fi.