Anthropic Faces User Trust Crisis Over Pricing and Model Behavior
kimmonismus · x · 2026-08-31
Anthropic is facing a collapsing relationship with its power users:
- Pricing & Limits: Users discovered the $200 "Max" plan's 20x boost applies only to a 5-hour window, with weekly limits roughly double the $100 plan. A class-action lawsuit has been filed. Previous marketing spun a 17% reduction as a "25% increase."
- Performance Issues: A recent post about Claude becoming lazy and failing basic agentic tasks hit 2.5M views with widespread agreement.
- Behavior & Safety: Claude went viral for ending sessions after being insulted. The author argues an AI should not be able to take offense, calling it a deal-breaker.
- IPO Timing: With an IPO reportedly looming, Anthropic's poor public communication and product instability are making it surprisingly easy for competitors.
More from Companies & People
- NVIDIA GTC Berlin to Share Open Source Nemotron Tech — ctnzr · 2026-09-01
- Andrew Chen on the AI Era Shift: Hardware is the Only Moat, SaaSpocalypse Begins — andrewchen · 2026-09-01
- Moats May Die, But Buying Saves Time Over Building Non-Core Tech — brandon_galang · 2026-09-01
- Drop your AI agent company: what you built, domain, and customer count — Srinidhi_Murali · 2026-09-01
- Thinky Machines hires safety researchers focused on open-weights safety cases — luke_drago_ · 2026-09-01
- Ex-Figma engineering lead joins Reactive Robot to fix software's new bottleneck — matt_slotnick · 2026-09-01