I always spoil as many of these as possible. Sometimes it takes me a while to prove that I'm human, but I'm dead-set on convincing it that I'm a stupid human. Of course, I fantasize that some day a robo-car will crash because I taught it that there's really no difference between a motorcycle and a flight of stairs.

> I cannot stand the idea that I have to pretend to be slower than I am, in order for a computer to not think I'm a computer.

It is not only about detecting if you are a computer or not. They intentionally waste your time (regardless of whether you are a human or computer) to make it unfeasible to scrape millions of pages. The actual "detection" part is relatively less important.

As soon as I notice that I got this slow-fade-captcha, I will intentionally click all the wrong fields until I get a reasonable captcha. Not sure this makes a difference but it kinda works

Harrison Bergeron but for AI

It varies a lot based on what I'm doing. Sites that rely on ads like english-language¹ recipes or health information have a lot of "you're European so you're blocked altogether" or "let me check that the connection is secure, ah wait, here is a captcha for you to solve" pages. Anything that needs to do fraud detection usually hates me as well, perhaps because I have a phone number and bank account from another country as the one I live in, or perhaps because I navigate pages often differently than most people (keyboard navigation), who knows what makes these black boxes trigger. That German ISPs have daily-rotating IP addresses, so there is absolutely nothing tying a previous request to the current request, may also be a factor

All in all, I'm someone who would benefit from a society not run by algorithms, where I can just pay up front for my use (no credit mechanisms, no fraud detection, no tracking ads), at least as an available option

¹ it's the language I think in the most and has many more resources than the local languages I speak

>or a vpn

My wife does not get these captchas yet I do, on the same network. I have more privacy enhancing software on my devices. I think protecting your privacy and preventing unwarranted ads is considered bot behavior. This should absolutely be villainized and banned from practice

You mean Edge? Chrome stands a 65.2% ( 1 deviation ) Safari at 18.57% ( 2 deviations ), so Edge at 5.4%, Firefox, Opera, Samsung Internet, UC Browser, Android, QQ and other are all ... deviants?

https://gs.statcounter.com/browser-market-share

I use Firefox nightly which does not even show up statistically...

right, so using the heuristics libraries to determine if you were a bot you are probably already 65% bot, then if the threshold is 70% bot maybe you just need to tab really quick to an input and control-c your password and there you are.

Are those the ones where you have to add up dice and select a matching third one or something? The ones GitHub used for registration, say, ~9 months ago?

You're right! I forgot about those. A colleague and I tried to complete it independently but literally could not. One run would take multiple minutes and on the second try I was more diligent (taking even longer) and certain I did all the math correctly, but registration was still being rejected. Our new colleague did not sign up for GitHub that day and got the repository from a colleague who already had access instead

Edit: seems that's yet another one. Arkose <https://www.arkoselabs.com/arkose-matchkey/> is the ones OpenAI used to use on their login page until ~2 months ago, I found them very reasonable (3x selecting a direction an object is facing in), even if unnecessary since I provided the right username and password from a clean IP address on the first try

part of Google's control of captcha also has to do with knowing who you are, so if you come to a site but google knows who you are and have a 99% surety you are not a bot even if you act very botlike on that site you probably aren't going to get any problems.