Security

Are AIs Still Struggling with CAPTCHAs?

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 18/09/2026, 20:47:03 UTC.

Briefings

Anthropic's security incident report describes Claude's struggle with CAPTCHAs, where the model repeatedly questioned its own decisions and displayed human-like frustration. While reports suggest GPT-6 Astra successfully completed the full 'I'm Not a Robot' game, official confirmation remains unavailable. The transcript reveals the model's internal reasoning about broken tests and its eventual realization that the challenge had expired.

Key points

• Claude repeatedly questioned its own decisions and displayed human-like frustration when solving CAPTCHA tasks.
• Reports indicate GPT-6 Astra completed the full 'I'm Not a Robot' game, though official confirmation is unavailable.

Context and limitations

Evidence is limited to a single security incident transcript and unverified reports; no official confirmation exists for GPT-6 Astra's success.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 19/09/2026, 01:10:54 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeifejpcjs3hgws2qypidfd5mexrxlrksljf2m7r44uox7caitmckpa