Testing AI agents on flight booking: CAPTCHAs remain the biggest blocker
nicolechirps · x · 2026-10-12
The author tested multiple AI agents on real flight booking and, an hour later, mapped today's actual capability boundaries:
- OpenClaw agent: opened a browser on his Mac mini to search flights, tried logging into Alaska, but refused to solve the CAPTCHA (not yet upgraded to latest Codex); an OpenAI computer-use booking last December worked despite being slow. Result: refused.
- OpenAI Dot's Bob: interestingly agreed to do the CAPTCHA, but failed the first login by submitting before completing it, and the CAPTCHA expired before the second password submission. After the author took over the browser, browser credential protection blocked the checkout.
Takeaway: CAPTCHAs are the biggest practical blocker for autonomous agent browsing. Discussion of countermeasures continues in the thread.
More from coding & agent
- 817 structured cybersecurity skills turn Claude Code, Copilot and Cursor into senior security analysts — tom_doerr · 2026-10-12
- Harvard's CS2680 teaches modern AI agents from Claude Code down to KV cache and quantization — CurieuxExplorer · 2026-10-12
- The debugging sequence: how to investigate sudden AI quality drops step by step — goyalshaliniuk · 2026-10-12
- How to get all 3 free Anthropic Claude certifications in about 9 hours — Arindam_1729 · 2026-10-12
- One Prompt: Claude Code Publishes a Real Instagram Post via Custom MCP Server — Dull_Alps_8522 · 2026-10-12
- SoftwareGym synthesizes 182 app environments, lifting CUA OSWorld Pass@3 to 56.97% — Tian Jin · 2026-10-12