Hands-on test: AI travel agents stumble on hotel booking

A developer's hands-on test tasking AI work assistants to plan a trip from emails found ChatGPT Work abandoning hotel booking midway and Claude failing on hotels too, with Gemini's feedback also flagged—showing agents remain unreliable on real multi-step tasks.

2026-09-16 ~ 2026-09-16 · 3 related posts

• Testing AI Work Assistants on Trip Planning: ChatGPT and Claude Both Fumble Hotel Availability — giffmana · 2026-09-16
• Test: ChatGPT Work gives up mid-trip-planning; Gemini flagged for wrong flights — DynamicWebPaige · 2026-09-16
• Agent tested on real trip-planning from email: ChatGPT nails trains, gives up on hotel availability — 3scorciav · 2026-09-16