Proposed framework for verifying AI agents' identity and permissions when acting on a user's behalf, akin to "know your customer" checks.
Baselayer's fraud consortium works because the same businesses keep applying across the 2,000 institutions on its network, and an AI agent that spins up for one task and disappears shows up once.
Reality
A dev.to roundup reads three agent papers around one design question: what a pending action still depends on at the moment it fires. The strongest number in it is 68 percent, and it belongs to somebody else's corruption test.
Reality
The Sao Paulo framework promises that an agent registered with one network will be recognized by the others, with certification and continuous monitoring attached. No specification, governance body or timeline has been published.
Reality
A New Stack prescription lists six identity capabilities for autonomous agents. The useful part is which old assumption each one retires, and what per-step credentials cost to run.
Reality
The report describes fraud as a chain, not a moment. That pushes spend toward verification and turns every channel where a human can be persuaded into a control point.
Reality