The Work Doesn't Stop

By Jackie Brock and Travis Guymon | Clearwater Global | July 1, 2026

Originally published on our Substack. Subscribe there to receive future articles by email. https://substack.com/@clearwaterglobal

One year ago, USAID was closed.

The programs ended. The needs didn’t.

USAID was not just a donor that left. It was connective infrastructure beneath the work itself: the evidence, the coordination, the supply chains, the market intelligence, the de-risking logic that let capital move into hard places at all.

A lot of people are operating as if that road is still there. It isn’t.

That matters to more than the people who worked inside the system. It matters to program teams trying to deliver, to local partners carrying more risk with less cover, and to funders and investors whose models were built on assumptions that no longer hold.

The risk did not disappear with the agency. It went invisible.

That is how this infrastructure always worked: it became invisible by succeeding. The oral rehydration therapy that is now the standard treatment for childhood diarrheal disease, responsible for cutting child deaths from diarrhea by more than half between 1980 and 2000, was developed in a USAID-funded lab in Dhaka. Nobody who administers it today names a U.S. government agency as the reason they know how. The substrate does not announce itself. It simply enables everything built on top of it. And that is why its removal is so hard to measure: the damage operates on a delay, and the instruments built to detect it are among the things going dark.

And the invisible part is what should worry people most. The damage is already measurable:

• Between 500,000 and 762,000 additional deaths in the first year alone. ¹

• 95 million people are losing access to basic healthcare.²

• 60,000 development workers displaced globally.³

• 23 billion in lost U.S. exports, and more than 200,000 American jobs at risk.⁴

The Lancet projects 14 million additional deaths by 2030 if the current gap is not filled.⁵

That clock started on July 1, 2025.

This is not a normal funding gap. The sector lost part of the substrate it had been standing on for decades. The people closest to the work can already feel it, even where the full consequences have not yet surfaced on paper.

That is why the usual responses feel thin. Better storytelling is not enough. More frameworks are not enough. Waiting for the center to move is not enough, especially if the center is gone.

So this is not a think tank. Not a commission. Not a position paper.

It is an operational answer to an operational gap.

That knowledge, human knowledge, carried by people, expressed in decisions, is the most valuable thing the development sector has right now. It didn’t disappear when the institution did. It lives in practitioners, in local partners, in the program officers, portfolio managers, and field coordinators who made judgment calls and lived with the results. And it is being wasted while everyone waits for someone to authorize the next move.

Clearwater Global is building decision-making infrastructure for people investing money, programs, and reputations in fragile and frontier environments. The logic is simple: AI can surface patterns, flag risks, and organize evidence fast, but humans still carry the call. Accountability does not disappear because a tool is involved. Human decisions need to be made by humans. That principle is not negotiable. It is the foundation on which everything else gets built.

That is why the work is built around decision-ready signals rather than decorative analysis. GO. STOP. CAUTION. WATCH. Clear enough for a program officer, a foundation director, or an investment committee to use when the facts are incomplete, and the consequences are not.

And the process has to be auditable. In one reviewed case, our system flagged fabricated source attributions and unsupported verification claims, blocking those signals from being treated as decision-grade until they were human-reconciled. That is the standard: evidence-bounded, challengeable, and accountable. Not replacing judgment. Giving judgment better instrumentation.

The people who know how to work in these environments are still here. The people who know which local partner can actually deliver, which assumptions in the model are already broken, and which risks are rising before they show up in a memo are still reachable. That knowledge doesn’t stay concentrated forever. If it disperses, the rebuild gets harder and slower.

We are not here to rebuild what was. We are here to make what comes next work better than what came before.

That is the conversation Clearwater Global wants to host. Not a conversation about what should exist in theory. A conversation among people who are still carrying real exposure now. Practitioners. Local operators. Funders. Investors. Critics with receipts.

Come argue with the line. Better yet, add to it. Bring the case the framework cannot price. Bring the proof. Bring the criticism. Find us at clearwater-global.com.

The work doesn’t stop. Neither do we.

Sources

¹ Center for Global Development, “Update on Lives Lost from USAID Cuts,” 2025–2026; ImpactCounter tracking model, January 2026 (762,000+ deaths attributed to USAID program cuts). https://www.cgdev.org/blog/update-lives-lost-usaid-cuts

² CP International, “The USAID Shutdown: Implications for Global Health and Humanitarian Aid,” 2025. https://cpintl.org/type/impact-story/the-usaid-shutdown-implications-for-global-health-and-humanitarian-aid

³ Oxfam America, “What do Trump’s proposed foreign aid cuts mean?” 2025. https://www.oxfamamerica.org/explore/issues/making-foreign-aid-work/what-do-trumps-proposed-foreign-aid-cuts-mean/

⁴ US Global Leadership Coalition, “International Assistance Cuts Hurt American Farmers, Workers and Businesses,” 2025. https://www.usglc.org/resources/international-assistance-cuts-hurt-american-farmers-workers-and-businesses/

⁵ Macinko J. et al., “Evaluating the impact of two decades of USAID interventions and projecting the effects of defunding on mortality up to 2030,” The Lancet, 2025. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(25)01186-9/fulltext