CAPTCHAs can still detect AI agents
Recent research indicates that while AI can solve CAPTCHAs, it does so differently than humans. The study highlights significant differences in the processes used by AI and human participants when completing these tasks. This suggests that AI's ability to mimic human output does not equate to a similar cognitive process.
- ▪AI can recognize images in CAPTCHAs but does not solve them like humans.
- ▪The study found statistically significant differences in error patterns and decision-making processes between humans and AI.
- ▪CogCAPTCHA30 was developed to measure the process behavior of humans and AI across various cognitive tasks.
Opening excerpt (first ~120 words) tap to expand
This is a ~1000 word overview of our recent machine learning conference paper submission. To read the full preprint, click here. "CAPTCHAs are broken these days." AI can easily identify all the traffic lights in a static grid. So CAPTCHAs don't provide a valuable human signal, right? Yes and no. Yes, because vision language models (VLMs) can recognize images like chimneys, fire hydrants, and traffic lights. Deep learning "solved" CAPTCHA-style image classification in the early 2010s. No, because AI does not complete CAPTCHAs like humans. If you look across all the data of humans and AI completing CAPTCHAs, you start noticing differences in features like error patterns.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Roundtable Research.