Built Bladerunner for #neatHack: an AI Cop that reads every agent trace in neatlogs, catches runs that break policy, proves it with two model families, auto-patches the guardrail with Entire provenance, and confirms the fix.
@neatlogs · @EntireHQ · @arithecfo
spent the weekend building a cop for AI agents lol
you run a swarn of agents, one goes rogue, Bladerunner catches it from the neatlogs traces and fixes its guardrails on its own
video made with opus 5.5
built for : #neatHack@neatlogs
masturbation addiction is crazy stuff, i remember one time i made a vow to stop, later that day i rubbed pepper on my hands so when the urge comes i’d remember i have pepper in my hands then i’d stop, you won’t believe the urge came 11pm that night and i went three streets away to get gloves.
and glory be to the devil, that day led me to the discovery of the best way to masturbate, gloves in your hands with lube all over it.
When the open-source AI platform Hugging Face was hit by a cyberattack this July, it quickly became clear this was no ordinary breach because the hackers weren’t human: they were autonomous agents created by OpenAI researchers.
Instadocs: AI Gone Wild reconstructs the breach as it’s never been seen before. Premieres October 12.