Why OAuth and Token Exchange Fail AI Agents: The Wrong Tool for the Wrong Job
·1740 words·9 mins
Understanding why forcing human-centric authentication patterns onto AI agents creates security theater instead of real accountability.
AI
There Is No State of the Art in AI Agent Authentication (And That's a Problem)
·1743 words·9 mins
On discovering we’re reinventing square wheels while ignoring centuries of working trust models—and what Red Hat can do about it.
When the weight of civilizations beats system design
·826 words·4 mins
A short essay on organisation darwinism: centuries of human coordination already shaped the patterns we now encode in protocols and law.
Power of Attorney Model: A civilizational approach for AI agent authentication, trust, delegation and revocation
·1519 words·8 mins
From JWT+SPIFFE fatigue to a Friday afternoon ban: borrowing Power of Attorney as a civilizational pattern for AI agents.