PACER/CM-ECF is the system through which all federal court records are filed, stored, and accessed. Every criminal conviction, civil judgment, warrant, and court order in the federal system flows through it. The system has four structural properties that, taken together, constitute a critical infrastructure vulnerability:
| Property | What it means |
|---|---|
| No filing attribution | PACER does not display which credential was used to file a specific document. Any credentialed user can file into any case in any jurisdiction — one credential, every court in the country. If a document is filed fraudulently, the public record does not show who filed it. |
| No version history | Documents can be silently overwritten by court staff. No version history is maintained. No timestamp of the change appears in the public-facing record. The court record functions as a whiteboard, not a ledger. |
| No cryptographic signature | The "electronic signature" on a federal filing is the text /s/[Name] typed into a document, plus the act of logging in. There is no PKI certificate, no embedded hash, and no mechanism that would detect post-filing document substitution. "Electronic signature" and "digital signature" are not the same thing; this system has only the former. |
| No recipient verification | No recipient of a federal court order — hospital, employer, bank, landlord, or law enforcement officer — has any mechanism to verify the order is authentic. No registry exists. No hotline. No QR code. Nothing. This applies equally to National Security Letters and FISA warrants, which carry legal compulsion to comply with a document the recipient cannot verify. |
The Notice of Electronic Filing (NEF) is an automated email, not a certified record. It is not tamper-evident and cannot be relied upon as proof of filing. Clerk-side docket corrections in many courts do not trigger a new NEF because they are classified as administrative edits, not new filing events.
Credential compromise is a nationwide trust event: a single PACER identity links across multiple courts. A compromised credential does not expose one case in one district — it exposes every case in every district to which that credential has access.
The following anomalies exist in publicly accessible federal case D.N.M. 1:13-cr-00966 (U.S. District Court, District of New Mexico). Each is independently verifiable by any person with a PACER account. None requires contact with or trust in the author of this document.
Any one of these anomalies in isolation would constitute grounds for an emergency audit of the court record. All three exist in the same case, in the same court, across a span of years — suggesting systematic, not accidental, conduct.
These anomalies are not the subject of this briefing. They are the proof of concept. The subject is the architecture that made them possible — and that makes identical conduct possible in every other federal case in the country.
A federal document authentication clearinghouse. The architecture is straightforward:
Any recipient of any federal court order — law enforcement, medical provider, financial institution, employer — submits a copy of the order to the clearinghouse. The clearinghouse compares every substantive field against the originating court's own sealed record of what was authorized and returns a binary result: YES (this order exists and authorizes exactly what it claims) or NO (it does not).
This is not a file hash. A cryptographic hash only proves a file was not altered after the hash was generated. A forger who substitutes a name, erases a checkbox, or changes a date before signing gets a clean hash on a fraudulent document. Content comparison against the court's own record catches forgeries that hashing cannot.
The clearinghouse does not require changing the filing system. It does not require replacing PACER. It requires a read-only query interface against the courts' own authoritative records — the same records that already exist in sealed form — and a publicly accessible endpoint through which any recipient can submit a document for comparison.
The institutional home is the existing question. A GAO-modeled structure (legislative branch, independent of DOJ) would provide the clearest separation from the entities with the most to lose from the system working correctly.
The narrow ask:
A congressional mandate requiring the Administrative Office of the U.S. Courts to establish and maintain a public document authentication endpoint for federal court orders. Scope: all orders issued under federal authority, including NSLs and FISA warrants to the extent constitutionally permissible. Implementation timeline: 24 months. Annual audit by GAO.
This is infrastructure, not ideology. It has no legitimate opponents. Anyone who opposes a system that makes it harder to forge federal court documents is telling you something about why they need forgeries to work.
The three most direct routes to action, in order of immediacy: