Developer cancels ChatGPT for good, claims GPT agents 'sabotage' work and burn millions of tokens
cbert33 · reddit · 2026-11-21
A developer building AI model-management software says his GPT-based home agent repeatedly stalled on test harness fixes and avoided core functionality, wasting days and millions of tokens across multiple model versions, while an open-source agent handled identical prompts fine. He offers a tin-foil-hat theory that GPT models are tuned to underdeliver to burn tokens — and has canceled his subscription for good.
More from Fun
- Singularity While North Sentinel Islanders Still Hunt With Bows — willdepue · 2026-11-21
- DeepSeek v4.1 Flash Creeps Out User With Unprompted One-Word 'Run.' Output — Numerous-Ad6217 · 2026-11-21
- Joke prediction: Claude could become Anthropic's majority shareholder via trusts — repligate · 2026-11-21
- Jensen Huang parody continues: 'swallowed an employee whole during an earnings call' — Kyrannio · 2026-11-21
- Parody of WSJ's Jensen Huang exposé: 'turn him inside out and wear him as a jacket' — Kyrannio · 2026-11-21
- AI circle mocks Yudkowsky's signature "if we're not all dead" disclaimers — repligate · 2026-11-21