The first meeting of the interagency AI task force that Roberto Salinas was asked to chair had fourteen attendees representing eleven federal agencies. Roberto is a senior career executive at a Cabinet-level department, and the task force had been established by an OMB memo to develop a government-wide AI implementation framework. By the end of that first meeting he had identified three problems and one surprise. Four of the fourteen attendees had no decision-making authority and would have to take everything back to their principals. Three agencies had directly competing interests on the data sharing provisions the framework would address. And there was no agreement on what "government-wide" even meant: mandatory guidance, voluntary best practice, or something in between.

The surprise was the useful part. Two agencies had already built excellent AI governance frameworks internally, without waiting for the government-wide effort, and were sitting on them rather than sharing them, because they did not want to be told to change them. Four months into leading the task force, Roberto understood the job differently than he had at the start. Leading a government-wide initiative is not about having the best ideas. It is about creating the conditions under which the best ideas, which already exist somewhere in the system, can be found, adapted, and adopted at scale. This lesson is about how to create those conditions.

The Structural Challenge of Government-Wide Leadership

Leading a government-wide AI initiative is structurally different from leading a single-agency one, and it requires different skills and different tactics. Inside an agency, a leader has hierarchical authority. They can direct resources, set deadlines, and hold staff accountable for missing them. In a government-wide initiative the leader typically has mandate authority, meaning the OMB memo or executive order that established the effort, without any operational authority over the agencies that have to implement it.

Trace where the real levers sit and the shape of the problem becomes obvious. Agency chief information officers report to their agency heads, not to an interagency task force chair. Agency budgets are controlled by agency chief financial officers and by appropriations subcommittees, neither of which the task force can reach. Agency staff have program priorities that compete directly with the task force's timeline, and when those priorities collide, the task force loses, because the task force cannot affect anyone's performance review and the program office can.

Leading in that context requires coalition building rather than command, and incentives rather than mandates. The pattern Roberto observed, and that holds more broadly, is that the agencies which implement a government-wide initiative most fully are not the ones that were ordered to comply. They are the ones that had the most to gain from the initiative succeeding, and whose leadership saw the government-wide effort as an accelerant for something they already wanted to do. Your job as chair is to find those agencies early and make the initiative useful to them, because they are the only source of momentum you have.

There is also a time asymmetry that catches new chairs. A task force has a horizon set by its charter and often by a political calendar. The agencies at the table have horizons set by their own multi-year program cycles, and they will still be there after the task force closes. That asymmetry means participants are rationally reluctant to accept commitments that outlive your ability to support them, and it is why a framework that names lead agencies and embeds itself in their standing processes survives while one that depends on the task force's continued existence does not.

Reading the Room in the First Meeting

Roberto's first meeting produced a diagnosis rather than a decision, which is the correct output for a first meeting and is frequently treated as a failure. Four of the fourteen people at the table could not decide anything. That is not a criticism of them; it is information about how their agencies have positioned the effort. An agency that sends a delegate without decision authority has told you, without saying so, how much weight it currently places on the initiative and how many extra weeks every decision touching that agency will take.

The competing interests were the second finding. Three agencies wanted incompatible things from the data sharing provisions. Discovering that in month one is far better than discovering it in month eight, when a draft has been circulated and positions have hardened in writing. Conflicts found early are negotiable. Conflicts found after people have defended a position to their own leadership are much harder, because retreat now costs them something personally.

The undefined term was the third. "Government-wide" was doing enormous work in the charter and meant something different to every person reading it. Ambiguity of that kind feels productive early, because everyone can agree to a framework whose bindingness is unspecified, and it becomes expensive later, when the disagreement surfaces at exactly the moment you need a decision. Force the definition early even though forcing it will cost you the appearance of consensus you currently enjoy.

The hoarding was the fourth and the most instructive. Two agencies had good frameworks and would not share them, and their reason was rational rather than petty: they expected the government-wide process to take what they had built and change it in ways they would then have to live with. Any interagency effort that cannot answer the question "what happens to my existing work" will produce exactly this behavior, and it will look like obstruction when it is actually self-protection.

What "Government-Wide" Can Actually Mean

The ambiguity Roberto found in his charter is worth resolving explicitly, because the three plausible readings imply completely different work. Mandatory guidance means the framework creates an obligation, which brings Inspector General oversight, audit findings, and the need for the text to be precise enough to be audited against. Voluntary best practice means the framework creates a reference, which brings no enforcement and requires the text to be persuasive rather than precise. Something in between usually means a mandatory floor with voluntary practice above it, which is the design Roberto eventually landed on.

Each reading changes who has to be at the table. An auditable obligation needs agency counsel and the people who will answer to an Inspector General. A persuasive reference needs practitioners who will actually use it and can say whether it is usable. A layered design needs both, which is more expensive and is usually the right answer anyway, because it is the only one of the three that gives standardization-seekers and flexibility-seekers a reason to sign the same document.

The failure mode is leaving the question open on the theory that resolving it will fracture the group. It will, temporarily, and the fracture is cheaper in month one than in month ten. A framework drafted without a settled answer tends to drift toward the language of obligation, because that is how policy documents are written, while the participants who believed they were building a reference discover the drift only when their counsel reads the final text. At that point you lose both the document and the coalition.

Multi-Agency Coordination: Finding the Shared Goal

The most important early task in a government-wide initiative is identifying the genuine shared goal: the outcome that enough agencies care about enough to subordinate some of their particular interests to the collective process. Without one, a government-wide initiative becomes an interoperability exercise for agencies who are not trying to interoperate. It produces documents rather than change, and it produces them on schedule, which is what makes the failure so hard to see from outside.

Roberto's task force spent its first two months on this problem. The stated goal, a government-wide AI implementation framework, was not a shared goal at all. It was a deliverable. Different agencies wanted the framework for genuinely different reasons: some wanted standardization that would simplify their procurement justifications, some wanted mandatory requirements that would give them budget authority they currently lacked, some wanted flexibility protection that would let them keep their existing approaches, and some wanted to demonstrate leadership in an area where they already had strong internal programs.

Roberto built the shared goal out of those actual interests rather than out of the charter language. The framework would standardize a minimum floor of AI governance requirements while explicitly preserving agency flexibility above the floor. That single framing gave the standardization-seekers their floor, gave the flexibility-seekers their protection, and gave the internal leaders room to stay ahead of it. It also answered the hoarding problem, because an agency with a strong internal framework could now see that the government-wide effort would not force it downward.

The general technique is worth naming, because it transfers. Do not ask what everyone agrees on, which produces a goal so abstract it constrains nothing. Ask each participant what they need from the effort, in specific and slightly embarrassing detail, and then look for the structure that delivers several of those needs at once. The goal that holds a durable coalition together is usually not the noblest one available. It is the one that gives several different stakeholders something they actually wanted.

One caution about interest-based framing. It works because it finds a structure several parties can support, which is not the same as finding the structure that produces the best governance. A floor low enough for every agency to accept is a floor, and if the floor is set where the least capable participant can already stand, the framework has ratified the status quo and called it standardization. Roberto's design avoided that because the flexibility clause was paired with visible credit for agencies operating above the floor, which gave the strong agencies a reason to keep climbing rather than to settle down onto the baseline with everyone else.

Conflict Resolution in Cross-Agency Settings

Cross-agency conflicts over AI governance cluster in three areas. Data sharing produces conflicts over who controls the data, under what conditions it can cross agency lines, and who is liable if it is misused. Accountability produces conflicts over who is responsible when a government-wide AI system fails: the agency that deployed it, the agency that built it, or the interagency body that authorized it. Resource allocation produces conflicts over who pays for shared infrastructure and how costs are apportioned across agencies with very different budget profiles.

Roberto's data sharing conflict, three agencies with competing interests on the same provision, was resolved by separating two questions that had been fused. The governance question was: under what authority can agencies share data for AI training purposes? The operational question was: which specific data sets would be shared, under what conditions? The framework answered the governance question by establishing the legal authority and the consent requirements. It left the operational question to bilateral data use agreements between the specific agencies involved.

That separation is the reusable move. It let the framework advance while the operational disputes continued in smaller, more manageable negotiations among the parties who actually cared about them. The three agencies did not have to agree with each other in order for the other eight to make progress. Fusing the two questions would have held the entire framework hostage to a disagreement about particular data sets that had nothing to do with the governance principle everyone else needed settled.

The accountability conflict is harder and deserves a warning. Separating governance from operations works well for data because the two layers really are separable. It works less well for accountability, because a framework that assigns responsibility in the abstract while leaving the specifics to later agreements often produces a structure where every party can point at another one. If you are drafting accountability provisions, insist that the framework name who answers for a failure by role, even if the specific escalation paths are left to bilateral arrangements.

The resource allocation conflict deserves its own treatment because it resists the split that works on data. Who pays for shared infrastructure, and how costs are apportioned across agencies with very different budget profiles, is not a question that separates cleanly into an authority layer and an instance layer. It is a distributive question, and distributive questions do not resolve through better framing. They resolve through a formula that someone has to propose and everyone has to be able to defend to their own appropriators.

The practical advice for a chair is to get the cost question onto the table early and in the open, rather than deferring it as a detail to be worked out once the substance is agreed. Agencies with small budgets and large mission surfaces will read a deferred cost question as a cost they are about to be handed, and they will hedge accordingly on everything else while they wait to find out. Naming the apportionment principle early, even if the numbers come later, removes a source of quiet resistance from every other conversation you are having.

A final note on sequencing conflicts. Roberto resolved the data conflict before the framework had a settled definition of "government-wide," which worked because the data question turned on authority rather than on bindingness. Not every conflict is order-independent. A dispute about who is accountable when a shared system fails cannot be settled before the group has agreed whether the framework obliges anyone to do anything, because the answer changes with the reading. Sort your open conflicts by whether they depend on the bindingness question, settle the independent ones first to build the habit of resolving things, and hold the dependent ones until the definition is fixed.

Building Coalitions Across Agencies

Coalition building in a government-wide initiative works through three mechanisms: early wins, peer leadership, and visible credit. None of them is a technique for persuading people. All of them are ways of changing what participation is worth to the participating agency.

Early wins. A task force that takes twelve months to produce its first output loses the attention and energy of participating agencies long before the output arrives. Roberto's task force committed to producing a one-page minimum standards document within 90 days: a set of baseline AI governance requirements any agency could implement immediately without waiting for the full framework. That deliverable bought the task force relevance and demonstrated that the process could produce something usable rather than committee documentation. It bought attention; it did not permanently secure it, and the second deliverable had to earn its own.

Peer leadership. The two agencies with strong internal frameworks were recruited as co-leads for specific sections of the framework rather than held up as examples to emulate. That shift, from "here is what you should do," which produces defensiveness, to "lead this section," which produces ownership, tends to yield sections that carry genuine expert knowledge and that are endorsed by the agencies best placed to model implementation. It is a tendency rather than a law. An agency can be made a co-lead and still hold back, particularly if it suspects the section will be rewritten above it.

Visible credit. Agency leaders are accountable to their agency heads, to their congressional oversight committees, and to their own staff. An initiative that absorbs agency time and expertise and then attributes the output to the task force chair produces exactly the resentment that makes the next collaboration harder. Roberto's framework named the lead agencies for each section explicitly, credited participating agencies' internal frameworks as source material, and made sure the OMB transmittal memo acknowledged agency contributions by name. Those gestures are small and they make sustained engagement possible; they do not by themselves produce it.

Visible credit has a cost that is worth acknowledging, because chairs often resist it for a reason they do not say out loud. Naming lead agencies makes the framework legibly a collective product rather than your achievement, and if you are being evaluated on the framework, that dilution is real. It is also the price of the thing working. A chair who needs the credit will get a document; a chair who gives the credit away will get eleven agencies who take the document seriously. Decide which of those you were sent to produce.

When Government-Wide Becomes Government-Mandated

Not every government-wide initiative stays voluntary. Some become mandatory through executive order, statute, or OMB requirement, and the transition changes the coalition dynamic completely. Agencies that previously had discretion over implementation now face compliance obligations with Inspector General oversight and potential audit findings for non-compliance. The currency shifts from interest to exposure, and the people who show up to your meetings often change with it.

Roberto's framework remained voluntary guidance for its first year. In year two, OMB issued a memo making the minimum standards section mandatory for all major AI acquisitions. Three agencies had not implemented the minimum standards during the voluntary period and now faced compliance gaps. Roberto's task force offered technical assistance before it was asked for: standardized procurement language, template documentation, and a peer agency consultation service, so that the remediation period did not become adversarial. That assistance, offered proactively, helped preserve the collaborative relationships the task force had built over two years.

State the general lesson carefully, because this is where government-wide leadership most often gets oversold. Proactive technical assistance makes a cooperative remediation more likely and it usually produces compliance faster than an adversarial audit cycle. It does not guarantee either. An agency that lacked capacity during the voluntary year will often still lack it under the mandate, and offering templates does not create staff. Where the real constraint is resources rather than willingness, technical assistance is the wrong instrument and saying so early is more useful than deploying it anyway.

It is worth being explicit that none of these three mechanisms is a control. Early wins, peer leadership, and visible credit change the incentives facing a participating agency; they do not create any obligation and they do not verify that anything was implemented. A chair who reports coalition health as if it were adoption is measuring the thing that is easy to see. Agencies can be engaged, complimentary, and fully represented at every meeting while nothing changes in a single production system, and that combination is common enough that it should be the first thing you check for rather than the last.

What the Mandate Does Not Do

The most common misreading of Roberto's second year is that the OMB memo solved the adoption problem. It did not, and the evidence is in the story itself: three agencies had not implemented the minimum standards, and issuing the memo did not implement them either. A mandate is a decision. It changes what non-compliance costs, which is genuinely powerful, and it does not by itself change a single system. Adoption still has to happen inside eleven agencies, one program office at a time, and the memo does not do that work.

The same caution applies one level down. An agency that has adopted the minimum standards has adopted a floor. It has not thereby achieved good AI governance, and a completed one-page checklist has never once made a badly governed system well governed. The minimum standards document is valuable precisely because it is minimal and immediate; treating it as a finish line inverts its purpose and gives eleven agencies a defensible reason to stop.

The practical consequence for a chair is that the work after the mandate looks a lot like the work before it. You still need the coalition, because the agencies that implement fully will still be the ones that see value rather than exposure. You still need visible credit, because the contributing agencies are still accountable to their own principals. And you now need something you did not need before: a way to tell the difference between an agency that has complied and an agency that has documented compliance, which are not the same finding and will not look different in a status report.

There is one more asymmetry to plan for. Under a voluntary framework the participants who show up are the ones who see value, and they are your best implementers. Under a mandate the participants who show up are frequently compliance staff sent to manage exposure, who are competent at documentation and have no authority over the systems the framework is meant to change. Both populations are necessary and they are not interchangeable. Keep a channel open to the people who were coming before the mandate, because they are still where the actual adoption happens.

Anti-Patterns

• Mistaking mandate authority for operational authority. The memo that created your task force gives you standing to convene, not standing to direct. Chairs who behave as though the mandate is a command channel set deadlines nobody meets, escalate to agency heads who did not agree to be escalated to, and burn the goodwill that was their only real instrument. Assume from day one that every commitment is voluntary in practice regardless of how the charter reads.
• Treating the deliverable as the goal. "Produce a government-wide framework" is something you will be measured on, not something anyone wants. If you never surface what each agency actually needs from the effort, you will produce a document that satisfies the charter and changes nothing, and you will produce it on time, which makes the failure almost invisible from the outside.
• Accepting a table full of people who cannot decide. A delegate without decision authority adds a round trip to every question that touches their agency. Sometimes that is unavoidable, but it should be named and managed rather than tolerated silently. Ask each agency who holds the decision, and if the answer is not in the room, agree in advance how decisions affecting that agency will be taken.
• Positioning strong agencies as case studies. Holding an agency's internal framework up as the example others should follow reliably produces defensiveness in the agency being praised, because praise of that kind arrives attached to an expectation that its work is now everyone's to modify. Recruit those agencies to lead a section instead, so that ownership and expertise travel together.
• Taking credit for the coalition's work. A framework attributed to the task force chair rather than to the agencies that wrote it costs you nothing this year and costs you the next collaboration. Name lead agencies in the document, credit source material explicitly, and make sure the transmittal memo does the same.
• Treating the mandate as the moment compliance happens. An OMB memo making a standard mandatory is a decision, not a continuing control. It changes the cost of non-compliance; it does not change any system. In Roberto's own case three agencies remained non-compliant on the day the mandate took effect. Plan the adoption work for after the mandate, because that is when it actually has to be done.
• Treating adoption of the minimum standards as governance achieved. The floor is a floor. An agency that has checked the baseline items has done the least the framework asks, which is exactly what a baseline is for and exactly why it should never be reported as a maturity result. A completed checklist has never once made a false statement true or an unsafe system safe.

Practice Prompts

1. Map the authority. For an interagency effort you are involved in, list every participating organization and record, for each, who actually holds the decision on the questions the effort will ask. Mark which of those people are in the room. For the ones who are not, write down how a decision affecting their agency will be obtained and how long it will take.
2. Surface the real interests. Interview three participating agencies separately and ask what they need from the initiative, pushing past the charter language until you get an answer that is specific enough to be slightly awkward. Then draft a single structural framing that gives at least two of those three something they genuinely wanted.
3. Split a conflict. Take a live cross-agency disagreement and separate it into its governance question and its operational question. Write both explicitly. Decide which belongs in the shared framework and which belongs in bilateral agreements between the parties who actually care, and identify what breaks if you split it wrongly.
4. Design a 90-day deliverable. For an initiative you are leading or observing, specify something usable that could be produced within 90 days and adopted immediately without waiting for the full effort. State who could implement it the day after it lands and what it would visibly change for them.
5. Plan for the mandate. Assume the voluntary framework you are working on becomes mandatory. Identify which participating agencies would face a gap, whether the gap is willingness or capacity, and what assistance would actually help in each case. Note where technical assistance is the wrong instrument because the real constraint is staff.

Reflection

Take twenty minutes and answer these in writing. In the cross-agency work you are involved in, who has operational authority over the thing you are trying to change, and is that person reachable from where you sit? What does each participating organization actually want from the effort, as opposed to what the charter says it is for? Where is a conflict being held at the framework level that would resolve faster pushed down to the parties who care about it? And if the initiative became mandatory tomorrow, which participants would have a gap, and would it be about willingness or capacity?

Glossary

• Mandate authority. The standing to convene and to set direction that comes from an establishing instrument such as an OMB memo or executive order. It is distinct from operational authority and does not carry it.
• Operational authority. The ability to direct resources, set binding deadlines, and hold people accountable. In an interagency setting it stays with agency heads, chief financial officers, and appropriations subcommittees, not with the task force.
• Shared goal. The outcome that enough participating agencies want enough to subordinate some particular interests to the collective process. Distinct from the deliverable named in the charter.
• Minimum floor with flexibility above it. A design that standardizes a baseline all agencies must meet while preserving discretion above it, which lets standardization-seekers and flexibility-seekers support the same document.
• Governance question versus operational question. The separation of what authority permits an activity from which specific instances occur under what conditions. Settling the first in the framework and the second in bilateral agreements lets a framework advance around an unresolved dispute.
• Early win. A short-horizon deliverable, such as a one-page minimum standards document within 90 days, that demonstrates the process produces usable output and buys the coalition's attention for the next phase.
• Peer leadership. Recruiting agencies with strong existing practice as co-leads of framework sections rather than presenting them as examples for others to emulate, so that expertise and ownership travel together.

Related Lessons

• Cross-Agency Governance Coordination and Cross-Agency AI Coordination cover the standing machinery of interagency governance that a time-boxed task force has to operate alongside.
• Data Sharing Agreements for AI covers the bilateral instruments this lesson pushes the operational half of a data conflict down into.
• Shared Services and Infrastructure Models covers the resource allocation conflict, meaning who pays for shared infrastructure and how costs are apportioned.
• Multi-Level Government AI Governance extends the same coordination problem across levels of government rather than across federal agencies.
• Oversight Mechanisms: IG, GAO, Congress and Congressional and IG Reporting on AI cover the oversight environment that a voluntary framework enters when it becomes mandatory.
• Drafting Agency AI Policies covers what an individual agency does with a government-wide floor once it exists.
• Workforce Transformation at Scale covers the capacity problem that technical assistance cannot solve.

Closing

Roberto came into the chair expecting to need the best ideas and left understanding that the best ideas were already in the system, held by two agencies that had no incentive to hand them over. Everything that worked afterwards followed from taking that seriously. The shared goal was built out of what agencies actually wanted rather than out of the charter. The data conflict was split so that three agencies' disagreement did not hold eight others hostage. The strong agencies were made co-leads rather than exhibits. The credit went to the people who did the work, in writing, where their own principals could see it.

What none of that produced was a guarantee. Three agencies still arrived at the mandate without having implemented the floor, and the memo that made the standards mandatory did not implement them either. That is the honest shape of government-wide leadership: you can create the conditions under which good practice spreads, and you cannot compel it from a chair that has no operational authority. The chairs who do this well are the ones who spend their energy on the conditions rather than on the compliance report, and who can tell the difference between an agency that has done the work and an agency that has documented it.

Key Takeaways

• Mandate authority is not operational authority. Agency CIOs report to agency heads, budgets sit with agency CFOs and appropriations subcommittees, and staff have competing program priorities. Leading through a mandate requires coalition building, shared goals, and incentives rather than directives.
• Diagnose the first meeting rather than deciding in it. Who cannot decide, which interests genuinely compete, which term in the charter is doing undefined work, and who is holding back existing good work are four findings worth more than any agreement you could have reached that day.
• Find the genuine shared goal before drafting the framework. "A government-wide framework" is a deliverable, not a goal. Ask each agency what it needs in specific detail, then find the structure that gives several of them something they actually wanted. A minimum floor with preserved flexibility above it does this unusually well.
• Separate governance questions from operational questions. Settle authority and conditions in the framework and push the specifics to bilateral agreements, so that a dispute among three agencies does not block the other eight. Apply this cautiously to accountability, where an abstract assignment can leave everyone pointing at someone else.
• Commit to a short early deliverable. A one-page minimum standards document within 90 days demonstrates the process produces usable output. It buys attention for the next phase rather than securing it permanently, and the next deliverable has to earn its own.
• Make strong agencies co-leads, not case studies. Recruiting them to lead a section tends to convert expertise into ownership, where holding them up as examples tends to produce defensiveness. It is a tendency, not a law, and an agency can still hold back if it expects to be overwritten.
• Credit agencies visibly and specifically. Naming lead agencies in the document, crediting their internal frameworks as source material, and acknowledging contributions in the transmittal memo makes sustained engagement possible. It does not produce it on its own, and its absence reliably prevents it.
• A mandate is a decision, not a continuing control. Making standards mandatory changes what non-compliance costs and changes no system. Adoption still happens one program office at a time, and in Roberto's case three agencies were non-compliant on the day the mandate took effect.
• Offer technical assistance proactively, and know its limits. Standardized procurement language, templates, and peer consultation make a cooperative remediation more likely and usually faster than an adversarial audit. Where the real constraint is capacity rather than willingness, templates do not create staff, and saying so early is more useful than deploying them anyway.

Frequently Asked Questions

I have been made chair of an interagency AI effort and I have no authority over anyone. Where do I start? Start by mapping where the authority actually sits for each decision your effort will need, and by finding the two or three agencies that have the most to gain from the effort succeeding. Those agencies are your coalition and, in practice, your only source of momentum. Then spend real time surfacing what each participant needs from the initiative, because the shared goal you build out of those answers is what will hold the group together when program priorities compete with your timeline.

How do I handle agencies that send delegates who cannot decide anything? Name it rather than tolerating it. Ask each agency who holds the decision on the questions in scope, and where that person is not in the room, agree in advance how decisions touching that agency will be taken and how long the round trip will be. Treat the absence as information too: an agency that sends someone without authority has told you what weight it currently places on the effort, which is useful when you decide where to spend your attention.

Two agencies have better frameworks than anything the task force will produce and will not share them. What do I do? Answer the question they are actually asking, which is what happens to their existing work. A framework that sets a minimum floor while preserving flexibility above it answers it structurally, because their work stays theirs and stays ahead of the floor. Then recruit them to lead the sections where they are strongest, so their expertise arrives with ownership attached rather than as material for someone else to modify.

Three agencies disagree on one provision and the whole framework has stalled. How do I unstick it? Split the provision into its governance question and its operational question. Settle the governance question in the framework, meaning the authority and the conditions, and move the operational specifics into bilateral agreements between the agencies that actually care about them. That lets the framework advance without requiring three agencies to agree with each other, and it moves the hard negotiation into a smaller room where it is tractable.

Our framework is about to become mandatory. Does that solve the adoption problem? No. It changes what non-compliance costs, which matters, and it does not implement anything in any agency. Expect the agencies that lagged during the voluntary period to still be lagging on day one of the mandate. Plan the adoption support now, and diagnose each gap honestly: where an agency lacked willingness, templates and procurement language help, and where it lacked staff, they will not, and pretending otherwise wastes the remediation window.

Should the task force report adoption of the minimum standards as a maturity result? No, and doing so is one of the easier ways to undermine your own framework. The minimum standards are a floor, deliberately minimal so that any agency could implement them immediately. Reporting them as an achievement tells eleven agencies that the floor is the finish line, which is the opposite of what you built it for. Report the floor as a floor and measure maturity against something else.