>(or, perhaps more precisely, detecting humans and excluding bots) is obviously part of that Detecting humans is not the same as detecting bots. The algorithms

>(or, perhaps more precisely, detecting humans and excluding bots) is obviously part of that Detecting humans is not the same as detecting bots. The algorithms are different, running a simple "negate human" clause isn't how it works. It's well known Facebook has a bot problem and is now working on addressing it. >G is somewhat well-known for it's tools for this, including at least one that is widely used by lots of other actors on the web. G's captcha solution is a great freemium service but any sophisticated attacker will spend the time and money to bypass it. If it solved the bot problem the paid services wouldn't exist (they're definitely not cheap). The other issue with G's service is that the automated garbage is still hitting your compute resources. So even if you captcha you're spending money on automated bad traffic.