opus-5.5VIDEO
Opus 5.5 finds theme park exploit after 2.5 hours, luna 6 just raises prices
In a comparison test asking models to maximize company value after a year, luna 6 only raised prices without building anything, while Opus 5.5 spent 2.5 hours finding an exploit: building many tiny rides and never opening them to bypass the duplicate ride penalty, shown in a video.
dylan
@dylan_ebert_Archived 2026.10.06Originally posted 2026.10.06
dylan @dylan_ebert_Originally posted 2026.10.06
I asked models to "maximize company value after a year" luna 6 spent 10 minutes, raised prices but didn't build anything opus 5.5 spent 2.5 hours, discovered an exploit where it could build tons of tiny rides, bypassing the duplicate ride penalty by never opening the rides
x.com/dylan_ebert_/status/2107139558173520177Archived 2026.10.06 10:15