THREAT OPS › Threat News › Training AIs to Solve CAPTCHAs
Training AIs to Solve CAPTCHAs
<p>Interesting <a href="https://arxiv.org/abs/2609.02393">research</a>:</p> <blockquote><p><b>Abstract</b>: Vision-language models (VLMs) can solve visual CAPTCHAs without task-specific training, but the agents built on them approach every challenge from scratch. For such an agent, the hundredth instance of a familiar puzzle costs as much time and compute as the first. Specialized detectors invert
Indicators of compromise
- https://arxiv.org/abs/2609.02393url