IARPA Funding Model vs. DARPA: Key Differences

IARPA — the Intelligence Advanced Research Projects Activity — is the Intelligence Community’s counterpart to DARPA: a program-manager-led research funder built on the same “ARPA model,” but sitting inside a different chain of command, serving a different customer base, and operating at a different scale. For research administrators and PIs evaluating whether an IARPA Broad Agency Announcement (BAA) is a fit, the mechanics look familiar if you’ve worked with DARPA — but the institutional context, transition pathway, and eligibility considerations are genuinely different.

What IARPA Is and Who It Funds For

IARPA sits inside the Office of the Director of National Intelligence (ODNI), not the Department of Defense. It was established in 2006 and began operations on October 1, 2007, formed by consolidating three existing intelligence-community research offices into one: the National Security Agency’s Disruptive Technology Office (DTO), the National Geospatial-Intelligence Agency’s National Technology Alliance (NTA), and the CIA’s Intelligence Technology Innovation Center (ITIC). Rather than building a new research-funding model from scratch, ODNI deliberately imported the DARPA model — IARPA’s first director, appointed in 2008, was a former DARPA program manager who consciously carried over DARPA’s high-risk/high-payoff approach.

IARPA’s stated mission is to “envision and lead high-risk, high-payoff research that delivers innovative technology for future overwhelming intelligence advantage.” Its transition customers are the elements of the U.S. Intelligence Community — agencies such as the CIA, NSA, NGA, DIA, and others — rather than the military services DARPA transitions to. IARPA funds no internal laboratories; every program is executed by external performer teams from academia, industry, and federally funded R&D centers, selected competitively.

The IARPA Funding Model

Structurally, IARPA runs on the same core mechanics as DARPA:

Term-limited Program Managers. Each IARPA program is defined and run by a government Program Manager on a fixed, multi-year appointment, not a permanent civil-service research staff. PMs design the technical goals, select performers, and can end underperforming efforts before a program’s planned close.
Broad Agency Announcements (BAAs) as the primary solicitation vehicle. New programs are announced through a BAA describing the technical problem, milestones, and evaluation criteria — the same instrument DARPA uses (see CASRAI’s guide to writing a DARPA BAA white paper for mechanics that largely transfer to IARPA BAAs as well).
Proposer’s Days. Before releasing a BAA, IARPA typically holds a Proposer’s Day where the Program Manager presents the program vision and a contracting representative outlines the award process — intended to help teams form before the formal solicitation drops.
Competitive, milestone-driven award structure. Programs are organized into phases with go/no-go milestones; teams that miss technical targets can be reduced in scope or cut before the program’s scheduled end, the same discipline DARPA applies.
Multiple award instruments. Depending on the specific BAA, IARPA has funded performers through contracts, grants, cooperative agreements, and, where authorized, Other Transaction (OT) agreements — the same family of instruments used elsewhere in federal research funding (see CASRAI’s overview of Other Transactions authority for how OT agreements work as a funding mechanism generally). The specific instrument for a given program is stated in that program’s BAA — don’t assume one applies across all of IARPA’s portfolio.

One point worth flagging for institutions unfamiliar with intelligence-community funders: IARPA has publicly stated that the majority of its research is unclassified and intended for open publication. That doesn’t mean every program is open to unrestricted foreign participation or free of security review — eligibility, personnel requirements, and publication constraints are set per-BAA — but it does mean “IARPA-funded” is not synonymous with “classified,” which is a common assumption among researchers who haven’t worked with the agency before.

How IARPA Differs From DARPA

Dimension	IARPA	DARPA
Parent organization	Office of the Director of National Intelligence (ODNI)	Department of Defense (Office of the Secretary of Defense)
Founded	Established 2006, operational October 2007	1958, in response to Sputnik
Origin	Consolidated three existing IC research offices (NSA DTO, NGA NTA, CIA ITIC) into one new activity	Built as a new standalone agency, no prior-office consolidation
Transition customers	Intelligence Community elements (CIA, NSA, NGA, DIA, and others)	Military services and combatant commands
Internal structure	A defined slate of named research programs reporting through the IARPA Director’s office, without DARPA’s standing multi-office model	Organized into permanent technical offices (e.g., Defense Sciences Office, Information Innovation Office), each running many concurrent programs — see CASRAI’s DoD research funding landscape guide
Scale	Substantially smaller budget and program count than DARPA; exact figures shift with each federal budget cycle	Multi-billion-dollar annual budget spanning a much larger program portfolio
Research classification posture	Majority of research is stated to be unclassified and openly published, alongside a smaller classified portfolio	Spans fully unclassified through highly classified defense work, depending on the office and program

What the two agencies share matters just as much as what separates them: both are built on the same “ARPA model” — external performers instead of in-house labs, term-limited program managers instead of permanent research staff, BAA-driven competitive solicitation, and a willingness to cancel underperforming programs early. That shared DNA extends further across the federal government: CASRAI’s guide to ARPA-E vs. DOE’s standard grant programs and the ARPA-H dictionary entry cover how the same model was later adapted for energy and health research respectively — IARPA was the first of that wider “ARPA-model family” to be built after the original DARPA, and remains the intelligence-focused branch of it.

What This Means for Research Administrators and PIs

If your institution is evaluating an IARPA opportunity, a few practical differences from a typical NSF or NIH proposal — and from DARPA specifically — are worth planning around:

Watch for a Proposer’s Day announcement before the BAA drops. IARPA, like DARPA, frequently uses these events to shape teaming before the formal solicitation is released; missing one doesn’t disqualify a proposal, but it does mean less lead time to form a competitive team.
Confirm the award instrument per-BAA, not by assumption. Because IARPA has used contracts, grants, cooperative agreements, and OT agreements across different programs, your institution’s contracts and grants offices should confirm which instrument a specific BAA uses before assuming standard federal-grant terms (e.g., Uniform Guidance under 2 CFR 200) will apply — contract-based awards carry different compliance and IP terms than grant-based ones.
Don’t assume classification requirements from the sponsor’s identity alone. Because IARPA sits within the Intelligence Community, institutions sometimes over- or under-estimate the security posture a given program requires. Check the specific BAA’s eligibility, personnel-clearance, and publication-review requirements directly rather than inferring them from IARPA’s parent organization.
Expect the same foreign-influence and research-security scrutiny that applies to other federal research security-sensitive sponsors. While IARPA’s specific disclosure and risk-review mechanics are its own, institutions already familiar with DoD-sponsor research-security review (see CASRAI’s DoD component decision matrix for foreign-influence risk review) should expect a comparable level of diligence, not a lighter one, simply because the sponsor is a research-funding activity rather than an operational IC element.

Frequently Asked Questions

Is IARPA part of the CIA or the NSA?

No. IARPA is a standalone research-funding organization within the Office of the Director of National Intelligence (ODNI), the office that coordinates the wider U.S. Intelligence Community. It was formed in 2006–2007 by consolidating research offices that had previously been run separately by the NSA, NGA, and CIA, but IARPA itself does not report to any single one of those agencies.

What is the core difference between IARPA and DARPA?

Both use the same program-manager-led, BAA-driven funding model, but IARPA sits within the intelligence community (ODNI) and transitions its research to IC agencies, while DARPA sits within the Department of Defense and transitions to the military services. IARPA is also considerably smaller in budget and program count, and does not organize itself into DARPA’s standing technical-office structure.

Can universities and small businesses without a security clearance work with IARPA?

Many IARPA programs are unclassified and open to academic and industry performers without pre-existing clearances, since IARPA has stated that most of its research is conducted and published unclassified. That said, clearance and eligibility requirements are set per-BAA, and some programs involving sensitive material do require facility or personnel clearances — always confirm the specific requirements in the BAA rather than assuming based on the sponsor.

Does IARPA use Other Transaction (OT) agreements like DARPA?

IARPA has funded performers through a mix of award instruments — including contracts, grants, cooperative agreements, and, where authorized for a given program, Other Transaction agreements. Which instrument applies is specified in the individual BAA rather than being uniform across IARPA’s whole portfolio, so confirm it program-by-program.

How does IARPA’s budget compare to DARPA’s?

IARPA operates at a fraction of DARPA’s budget and program count. Precise current figures move with each federal budget cycle and are best checked directly against ODNI’s published budget justification materials rather than assumed from older figures.

IARPA: Funding Model and How It Differs from DARPA