A clean connector does not make a reliable agent
Connector correctness and agent behaviour are separate problems. Most teams only test the first one.
#ai-safety
Posts tagged #ai-safety
Day 58: Why governance belongs in the architecture
Why governance belongs in the architecture, not in a document after launch.
Day 57: Why retrieved content must stay untrusted
Why retrieved content must stay data, not become authority over the system.
Day 52: Why shared memory becomes a permission problem
Why shared memory becomes an access-control problem in multi-agent systems.
Day 38: Why guardrails should enable safe action
Why guardrails should shape safe progress, not just block uncertain behavior.
Day 37: Where human review actually belongs in AI workflows
Why human review needs timing, authority, and context to be useful.
