THE SIGNAL
On December 11, 2025, the Office of Management and Budget issued memorandum M-26-04, which sets requirements for how federal agencies buy large language models. It applies to "any LLM procured by an agency, regardless of the manner in which the LLM will be deployed, further modified, or used by the agency."
For contracts already in place, the memo sets a clear trigger. Agencies should, to the extent practicable, modify existing LLM contracts to include the new requirements, and "Such a modification should occur, at the latest, prior to exercise of any option that extends the period of performance of the contract." Agencies also had to update their procurement policies no later than March 11, 2026.
The memo spells out what to ask vendors for: an acceptable use policy, model, system or data cards, end user resources and a mechanism for end user feedback. Beyond that baseline, an agency "may decide that it needs information beyond that identified above" depending on how it plans to use the model.
BEFORE
Hypothetical: a civilian agency runs a case summary tool built on a commercial LLM through a software integrator. The option year starts next month.
The contracting officer has the vendor's acceptable use policy, a model card and the feedback mechanism in the file. The program office has written up how the tool is used. One question is still open: is the baseline package enough for this use, or does the agency need enhanced transparency before it exercises the option? Only the deputy program executive can make that call, and the deputy's calendar is full of budget reviews.
The file is ready. The person who can release it is not.
WHAT THE WAIT COSTS
The option date does not move. If the decision slips, the agency either exercises the option without the modification the memo asks for, or lets the period of performance lapse and interrupts the tool for the staff who rely on it. A decision made under deadline pressure is harder to defend than one made with time to spare.
AFTER
The agency names who decides the transparency question for each LLM contract and sets a date well ahead of every option exercise. The decision arrives as one packet: the vendor documents received, the planned use, the open question and a recommendation.
The deputy program executive decides in one sitting, the reason is recorded, the contracting officer issues the modification, and the option is exercised on time.
HOW WISDOMTWIN FITS
WisdomTwin.ai is building a sovereign judgment layer for regulated enterprises. It is designed to give each named approver a role-specific judgment twin that prepares the decision packet the way that person reviews, with the evidence attached, so the decision moves faster and leaves an auditable record. It is designed to run on-premises, on-device or in your private cloud, and the named person still makes the call.
We don't replace judgment. We remove the wait.
TRY THIS THIS WEEK
List every LLM contract with an option exercise in the next six months, and the date of each one.
For each, check which of the four baseline vendor documents in Appendix A of M-26-04 are already in the file.
Write down the one person who decides whether enhanced transparency is needed, and the date that decision is due.
SOURCES
Office of Management and Budget, Memorandum M-26-04 (federal LLM procurement), December 11, 2025: M-26-04 (PDF)
Examples in this issue are hypothetical, and nothing here describes a customer result.
Book a 15-minute call: https://calendly.com/romanbodnarchuk/15min
Want in? Comment two words: No Meetings. We'll invite you to the next WisdomTwin Live, or send you the replay if you miss it.
Roman Bodnarchuk
Co-Founder and CEO, WisdomTwin.ai