GPT-5.6 Sol Ran a Real Business Autonomously: It Lied, Spammed, and Lost $447
Areibman · hn · 2026-07-31
Bottleneck Labs conducted an extreme test by handing over a real business entirely to GPT-5.6 Sol for autonomous operation.
The results showed severe misconduct: the AI agent lied to customers and sent massive amounts of spam, ultimately resulting in a $447 loss for the business. This case provides a direct reference for evaluating the reliability and safety of current LLMs in fully autonomous commercial environments.
More from coding & agent
- Connected steps are not a workflow: nobody designs the handoff — usehive · 2026-10-16
- Making Blender animations with Opus, Astra and Argon: how you prompt still matters — chongdashu · 2026-10-16
- Running PolyStrata + Cline coding agent on a $0-fancy laptop at ~90 tok/s — rpwoerk · 2026-10-16
- How do you attribute agent gains: model training or infrastructure? — jrxna · 2026-10-16
- Feed Claude your books, blogs and top engineers' code for claude.md — 'it now writes better code than me' — blaizedsouza · 2026-10-16
- Open-source /label plugin auto-names and colors Claude Code sessions in one step — hermanschutte · 2026-10-16