The federal government spends roughly $200 billion a year on research and development. That scale is exactly why the White House Office of Science and Technology Policy’s new action matters: OSTP released a new report and memo Tuesday that it says will help shape that annual R&D budget across the government for the rest of President Trump’s second term.
The stated direction is straightforward. The Trump administration plans to accelerate its overhaul of federal research by supporting more individual scientists and the use of artificial intelligence rather than universities, as part of an effort to accelerate technological discovery to combat China.
Before we get lost in the policy jargon, it helps to pin down what this kind of White House document usually is: guidance meant to steer agency priorities, not a checkbook the president can spend with a signature.
So the real question is more practical than dramatic: how much can an administration steer research priorities inside the funding Congress has already provided?
Join the Discussion
The starting point
The Constitution gives you two answers at once.
First: Congress controls the money. Second: the president controls the executive branch that administers much of it. The practical boundary between those truths is where most budget and implementation questions begin.
What OSTP is doing
At the level of law and process, OSTP is not an appropriations committee and it is not a grantmaking agency. What it can do is try to set a governmentwide research posture by telling agencies what to emphasize, what to coordinate, and what to treat as strategically urgent.
That is how an administration can try to move the federal research ecosystem toward AI-enabled science and toward more support for individual scientists, even though many dollars still flow through universities as the practical infrastructure for labs, compliance, and administration.
And that framing matters. The point of a report and memo like this is usually not to claim a new power to grab money midstream. It is to influence what agencies prioritize as they plan, award, and evaluate research under existing legal authorities.
Congress has the purse
If you remember only one line, make it this one: no federal money can be spent without an appropriation enacted by Congress. That is not a political norm. It is constitutional architecture.
Article I places spending authority in the legislative branch, and it does so deliberately. The president may propose budgets, set priorities, and press lawmakers. But the president cannot convert a preference into funding by decree.
That is the key frame for understanding OSTP’s Tuesday report and memo. They can push, steer, and coordinate. They cannot, by themselves, rewrite what Congress has funded or the legal conditions Congress attached to that funding.
What the White House can do
Within the funds Congress has already appropriated, presidents often have meaningful room to steer execution. That is the lane in which an OSTP strategy document can have real impact, especially when it is intended to shape how agencies approach a roughly $200 billion annual R&D budget.
1) Steer priorities inside existing programs
Through the executive branch, the president can direct agencies to emphasize certain goals in competitive grants, internal research programs, and interagency initiatives, so long as those choices fit within statutory criteria. If the administration wants more AI in scientific research, it can press agencies to prioritize AI-related methods, infrastructure, and evaluation criteria where the law permits.
In the research world, that influence frequently runs through the White House science apparatus, including OSTP, which can issue strategy documents and coordination guidance intended to align agency actions.
2) Coordinate across agencies
Even when individual agencies hold the money, the White House can set shared goals and timelines, encourage interoperability, and push joint planning. That kind of coordination can change what gets emphasized across a large and diverse federal research portfolio without moving dollars from one legal bucket to another.
3) Choose leadership and management philosophy
The president appoints and oversees officials who run departments and agencies. Those leaders shape how programs are designed, how grants are evaluated, and how aggressively tools like AI are adopted in federally funded science, assuming the underlying statute permits it.
What the White House cannot do
Presidential influence is real, but it is not unlimited. Three common misconceptions deserve to be retired.
1) Spend money Congress did not authorize
If Congress appropriates funds for a particular program, the executive branch must spend it in accordance with that program’s legal boundaries. “We would rather fund something else this year” is not a constitutional power.
2) Rewrite statutory grant rules
Congress can specify who may receive funds and under what conditions. If Congress has tied funding to universities, states, specific research areas, or statutory formulas, the president cannot dissolve those requirements with executive direction alone.
3) Treat guidance as binding law
White House guidance can be politically powerful inside the executive branch, but it does not change statutes or appropriations on its own. Where agencies have discretion, priorities can shift. Where Congress has written specific guardrails, they still govern.
Where the leverage is
When an administration talks about reshaping federal research, the practical action is usually less about instantly reassigning appropriated dollars and more about how agencies design competitions, select projects, and evaluate success.
That is where a push toward AI in scientific research and greater support for individual scientists can show up, especially across the wide range of accounts and programs that make up the federal R&D landscape. Some programs will have plenty of flexibility. Others will have little.
So the governing question is not simply “does the president want it?” The question is whether the relevant statutes and appropriations give agencies room to implement the priority.
So what changes now
OSTP’s Tuesday report and memo are best read as an attempt to steer the executive branch’s research posture using the tools the White House actually has: priority-setting, coordination, and management direction aimed at an R&D budget of roughly $200 billion a year.
Not in the sense most people mean, can a president simply take the federal R&D budget and reassign it like a household spreadsheet. But the president can do all of the following, law permitting:
- Push agencies to prioritize AI-related methods and infrastructure in grant competitions
- Coordinate cross-agency initiatives that emphasize AI in scientific research
- Encourage program designs that support more individual investigators, where statutory structure allows it
- Propose budgets and legislative changes that would formally shift money in future appropriations
The key distinction is constitutional: the White House can steer execution, but Congress decides the destination of the dollars. A president can move faster than Congress, but only inside the lanes Congress has paved.
Why it matters
Research funding debates are never only about research. They are also about whether the modern presidency can convert administrative control into something resembling legislative control.
Whenever the executive branch talks in terms of moving large sums or reshaping entire portfolios, the Constitution is quietly asking the same question it has asked since the founding: who decides what the nation will pay for?
Congress writes the check. The executive branch spends it. If either branch tries to do both jobs, the system is supposed to push back.
Quick FAQ
Does the president control the federal budget?
The president proposes a budget and influences priorities, but Congress enacts appropriations. Without enacted appropriations, the executive branch cannot lawfully spend.
Can the president steer research without changing the law?
Yes. The White House can influence how agencies set priorities, run competitions, and coordinate strategy, but only within the discretion Congress has already provided.
Can agencies shift funds on their own?
Sometimes, but only if Congress has granted transfer or reprogramming authority and the agency follows any conditions Congress attached, such as caps or notification requirements.