Ethan Mollick, Wharton associate professor, uses an autonomous GPT agent to win a Slay the Spire 2 daily challenge
The model operated with full computer control for five hours.

This was one of those impressive AI thresholds for me. I gave GPT-5.6 Sol in Codex control over my computer, and asked it to win the daily challenge for the game Slay the Spire 2 (randomized factors, so can't cheat). It worked for 5 hours, making complex game choices... and won.