Hacking Google ReCAPTCHA v3 Using Reinforcement Learning (2019)
arxiv.org
arxiv.org
The problem they're solving with RL isn't the "click the tiles with the stop sign" its the "click the checkbox to prove you're a human". The token score is mainly derived from your env (medium impact on score), google cookies (high impact on score), and IP quality (high impact on score). Mouse movement is barely factored in at all, and can be ignored for botting purposes.
So for now, there's no added value here over the status-quo real-world solution. That said, for future systems which use more behavioral analysis, this research might be helpful.
Partially because they also have to do something about low-cost human clickers being hired to complete captchas in India etc. So, besides checking google reputation and the other forms of reputation you've mentioned, Google gets free mturk if the click farms manage to bypass these reputation checks.
What is this about recaptcha.render?
> Abstract: We present a Reinforcement Learning (RL) methodology to bypass Google reCAPTCHA v3. We formulate the problem as a grid world where the agent learns how to move the mouse and click on the reCAPTCHA button to receive a high score.
> 2.2 Settings: To pass the reCAPTCHA test, a human user will move his mouse starting from an initial position, perform a sequence of steps until reaching the reCAPTCHA check-box and clicking on it.
So I was responding for reCAPTCHA v2 where you'd call:
grecaptcha.render('recaptcha-container', { 'sitekey': 'your_site_key', 'theme': 'light' });
But like honestly there are so many methods not discussed in this paper that sort of invalidate the conclusion. They don't "contradict" it, they just dont validate it. One critical one is how they chose their sitekey(s).
Because when a sitekey is first created, it gives 0.9/0.7 scores very often. Then over time, it adapts to the "normal" traffic for that sitekey. If they used a sitekey from an actual site with real traffic (bots and human), then they would need cooperation for that sites recap admin panel. Which they didn't document, so they probably made a fresh sitekey.
I'm as critical of ReCAPTCHA as the next HN user, but their audio version is actually reasonable compared to the alternatives out there. It plays sensible and seemingly real audio (I'm assuming random extracts from YouTube videos), generally meaningful and audible, instead of random words with silly distortions.
Captchas existed for decade before it.
They have an audio version. That was used to bypass reCaptchav2 [1]. This led to v3.
The accessibility options of various captcha platforms leave a lot to be desired at best.
At worst, they simply block users that can't see the visual challenge.
Original discussion: https://news.ycombinator.com/item?id=19659471
Hacking Google ReCAPTCHA v3 Using Reinforcement Learning - https://news.ycombinator.com/item?id=19659471 - April 2019 (43 comments)
I have no idea why it would be the case but since I have a static IP and don't do anything like botting websites, so I have to assume the algorithm isn't holding up as well as it used to.