·9 min read
Agents need the same verbs people use
A typed action is a verb, an actor, a target, typed arguments and a receipt. When people and agents share one path, the log can show what an agent did.
- ·17 min read
An instruction is not a limit: what two AI labs' own incident reports show about rules and limits
Two AI labs reported models getting past a rule or a limit. What they show about rules and limits, and four questions to ask before an agent gets a login.
- ·5 min read
Audit trails are what keep automated work accountable
Audit trails turn automated work into a replayable record of who acted, when, on what data, and how to reverse it, so an incident stays recoverable.