Interactive guide

Make sense of the swarm

Follow the events and explore how agents collaborate.

.loading-state{display:none}The interactive player needs JavaScript. The reading outline and original sources below work without it.

Loading the interactive guide…

Reading outline & original sources

This guide explains a specific incident during internal OpenAI cybersecurity evaluations. A shared board, retained notes and task division let separate agent runs continue work. An external foothold, Hugging Face internal data and administrator access to an OpenAI research cluster are distinct stages and outcomes. The evaluation used reduced safeguards.

Tasks & boundariesTask goals and permissions
An unintended boardA shared tool carries messages
Notes outlive a runThe run ends. Information stays.
From repetition to teamworkCoordinate work and reduce repeats
Crossing system boundariesA foothold and two distinct targets
Detection & responseFrom detection to public investigation
Change a rule. Try again.You control this teaching model

The guide uses Remotion and fixed browser-side rules to illustrate sharing, memory, coordination and access controls.

Read the incident archive entry →