Cryptographic integrity is not measurement integrity. An AI agent can sign a receipt for a check that never ran, measured the wrong thing, or covered only part of the claim.
Your AI agent returns a signed receipt:
0 defects found.
The signature is valid. The receipt has not been altered. The agent was authorized to run the check.
The result can still be wrong.
Perhaps the scanner hit a rate limit and silently converted eleven failures into eleven empty results. Perhaps a watchdog inspected 8 machines and issued a conclusion about 68. Perhaps a database health check ran select 1 successfully while the application was failing because a required column did not exist.
I built a hostile AI agent receipt that passed the cryptographic checks. The signature was valid....
A hands-on demo of cryptographically verifiable agent behavior — a tamper-evident receipt proving an agent ran the approved…
This post builds on a conversation that started in Ian Johnson's harness stack thread and the...
I have been writing here about one core idea: AI agents do not only fail because they forget...
An AI agent's report is not evidence, and a tool's success message is not evidence either. What counts as proof, the five checks…
Attach a cryptographically signed receipt to every LLM call so you, your auditors, and your customers can verify what your AI…
