AER-1 is an open standard (IETF draft-zambo-aer1-26) defining a verifiable receipt for AI agent work: a tamper-evident record that proves the saved result of a job was not changed after the fact. The project ships the standard text, a frozen conformance corpus (143 test vectors, 146 checks), reference verifiers anyone can run for free, and public tooling such as the Receipt Checker and the Claims-vs-Receipts Checker. Five independent implementations are conformant today and an independent validator re-verified the corpus at 119 of 119 checks. Funding keeps the corpus, verifiers, and specification maintained as free public infrastructure and supports the next independent implementations through conformance testing.