User switches from Claude to ChatGPT citing quality decline and strict quotas
warlordthe99th · reddit · 2026-08-22
A long-term Claude subscriber explains why they are switching to ChatGPT. The user reports that Claude 3.5 Sonnet (Opus) has regressed in creative analysis and abstract problem solving, even performing worse than free models in some tests. They also criticize the strict usage limits (5-hour caps) for blocking productivity during peak hours. Furthermore, the post critiques Anthropic's corporate stance and warns that upcoming text watermarking features could severely impact coding workflows.
More from Companies & People
- University of Toronto hiring AI postdoc: builders of interventions, not just benchmarkers, due Oct 30 — ishtiaqueSIA · 2026-10-16
- Fluidstack hires Electrical Engineer for Modular R&D to build frontier AI infrastructure — MxMnr · 2026-10-16
- Databricks engineers land impactful PRs on day 2 with AI dev tools, vs week 3 before — Yuchenj_UW · 2026-10-16
- Jensen Huang joins BG2 podcast: does AI revenue justify capex, '27 bottlenecks — elonmusk · 2026-10-16
- Local media growth via Meta ads: UGC-style creatives win, exclusion targeting matters — TJLarkin23 · 2026-10-16
- Anthropic grants METR employee-like access as independent safety evaluators go public — AnthropicAI · 2026-10-16