When Deputy Secretary of Defense Kathleen Hicks unveiled Replicator at an August 2023 defense conference, she set a specific, public benchmark: field attritable autonomous systems “at scale of multiple thousands, in multiple domains, within the next 18-to-24 months.” By late 2025, the Pentagon reported “hundreds” delivered. That shortfall is not simply a procurement miss. It exposes a structural problem the initiative was designed to solve but has not yet fully solved: the U.S. defense industrial base was not configured to produce expendable autonomous systems at the pace and volume that Ukraine’s war has demonstrated are operationally necessary.

The August 2023 Announcement: Mass Against Mass

Hicks announced Replicator on August 28, 2023, at the NDIA Emerging Technologies for Defense Conference in Washington. The threat framing was direct.

“Replicator is meant to help us overcome the [PRC]’s biggest advantage, which is mass. Counter the [PLA’s] mass with mass of our own, but ours will be harder to plan for, harder to hit, and harder to beat.” — Kathleen Hicks, Deputy Secretary of Defense

The proposed solution was a deliberate inversion of the Pentagon’s historical procurement logic. Rather than platforms that are large, exquisite, expensive, and few, Hicks argued for systems that are “small, smart, cheap, and many” — attritable assets cheap enough to absorb battlefield loss without stranding an entire capability, and procurable fast enough to sustain high-attrition operations. The contrast was not rhetorical; it was a direct critique of a procurement culture that optimizes for longevity over mass.

The Ukraine precedent was central to the argument. By mid-2023, Ukrainian forces were losing large numbers of UAS each month — to Russian electronic warfare, air defenses, and kinetic intercepts. The operational lesson was legible: drones deployed at scale must be cheap enough to lose, numerous enough to complicate any defensive response, and procurable fast enough to maintain inventory under attrition. Hicks specified a three-to-five year intended lifecycle — a direct signal that the acquisition model needed to change, not just the hardware. She also drew a deliberate line on autonomy: these systems were “not synonymous” with weapons, and humans would retain use-of-force decisions throughout.

Architecture: Existing Money, New Acquisition Velocity

Replicator’s structure was engineered to bypass the slowest features of traditional defense acquisition without requiring new legislation. It was not established as a program of record. The department did not request new FY2024 appropriations; instead, it consolidated existing service programs under new coordinating authority and leveraged existing spending authorities. That approach allowed faster action but made congressional oversight structurally harder — a problem that would resurface explicitly in CRS reporting two years later.

Governance was anchored at the department’s senior leadership tier. A Deputy’s Innovation Steering Group, co-chaired by Hicks and Vice Chairman of the Joint Chiefs Admiral Christopher Grady, oversaw execution. The Defense Innovation Unit served as the implementing agency, with DIU Director Doug Beck on the working group. The primary acquisition vehicle was DIU’s commercial solutions opening (CSO) mechanism — a contracting pathway designed for rapid engagement with non-traditional defense vendors that bypasses much of the traditional source-selection timeline. Maritime systems were solicited through the PRIME CSO, which drew more than 100 applications.

Funding came from within existing appropriations: approximately $300 million from FY2024 appropriations, rising to roughly $500 million for the full first tranche when existing authorities were folded in. FY2025 requests ran at a comparable level. The decision to finance through existing authorities rather than a new budget line gave the initiative speed and flexibility. The cost was visibility: without a distinct appropriation, oversight bodies had no single number to track, and the public cost picture remained thin throughout the program’s first two years.

Tranche 1 and Tranche 2: Named Systems, Classified Gaps

The first tranche (Replicator 1.1, disclosed May 2024) named four vendors publicly: AeroVironment’s Switchblade 600, Anduril’s Altius-600 and Ghost-X, and PDW’s C-100, along with several unnamed software vendors. Maritime systems were procured through the PRIME CSO; some maritime and counter-UAS capabilities remained classified.

The Switchblade 600 and the Altius-600 are both one-way loitering munitions — the category that proved decisive in Nagorno-Karabakh and has since become a defining feature of attritional drone warfare in Ukraine. Their inclusion placed Replicator squarely in that mission domain. The Ghost-X is a multi-mission UAS platform suited to longer-endurance ISR and targeting roles rather than one-way strike. The several unnamed software vendors point to a parallel investment in autonomy and resilience software that the department treated as equally important as the physical platforms.

Tranche 2 (Replicator 1.2, announced November 21, 2024) broadened scope to additional air and maritime platforms, software autonomy and resilience enablers. DoD set an August 2025 deployment deadline for Tranche 2 systems.

The Counter-UAS Pivot, the Delivery Gap, and an Unresolved Industrial Question

In September 2024, Secretary of Defense Lloyd Austin announced Replicator 2, reorienting the initiative’s second phase toward defeating small UAS threatening fixed installations. The immediate cause was Tower 22: a January 28, 2024 drone strike on a U.S. base in Jordan that killed three American service members and injured dozens more. Austin framed the new mission as addressing “the warfighter priority of countering the threat posed by small uncrewed aerial systems to our most critical installations and force concentrations.”

The cost-exchange arithmetic that drove the pivot is hard to dispute. Iran’s April 2024 combined-arms attack — ballistic missiles, cruise missiles, and one-way UAS — cost an estimated $80–100 million to mount. The Israeli and allied defensive response cost approximately $1 billion to execute. Defeating cheap drones with expensive interceptors is arithmetically unsustainable as a baseline force posture, which is the central problem that layered counter-UAS architectures are built to address.

Replicator 2.0 — a follow-on solicitation issued by DIU alongside NORTHCOM and the Joint Counter-small UAS Office in May 2025 — specified low-collateral-effects counter-sUAS for fixed installations, with a deadline of May 16, 2025. The solicitation was explicitly aligned with a Trump administration executive order on acquisition modernization issued April 2025, suggesting cross-administration continuity on the underlying urgency even as some implementation priorities shifted.

The delivery picture as of late 2025: T.S. Allen, the former Replicator 1 director, stated the program had supplied “hundreds of drones to warfighters” with “thousands more on contract…still rolling off the assembly line.” That formulation is not a retraction of the original goal, but it is a significant distance from the 18-to-24 month timeline for fielding “multiple thousands” that Hicks announced at the program’s launch. The gap reflects real industrial constraints: American UAS manufacturers were not capitalized or staffed at the production rates the initiative requires, and accelerating to those rates takes time that a two-year deadline did not leave.

The Congressional Research Service documented the oversight dimension in report IF12611. Public cost information remained minimal. INDOPACOM raised concerns about whether range and weather conditions in its theater matched the characteristics of the systems Replicator had selected. And the initiative’s cross-authority financing made systematic tracking difficult for Congress and outside reviewers alike. CRS recommended mandated reporting requirements and GAO review authority over the program. The American Enterprise Institute, in congressional testimony, identified a different risk: that sustained Replicator spending could crowd out “funding for near-term munitions” — potentially trading existing combat-ready stocks for autonomous systems that were still scaling up production.

DIU Director Doug Beck resigned in August 2025. Beck had been a central architect of Replicator’s structure and the most prominent internal advocate for its acquisition model; his departure introduced continuity uncertainty at a moment when the program was mid-execution across two tranches and transitioning toward a new counter-UAS focus under Replicator 2.

The honest read of Replicator’s first two years is that it correctly identified the problem — U.S. acquisition machinery was not built for expendable autonomous systems at the pace Ukraine-scale operations demand — and created a real, if imperfect, structural workaround. The CSO mechanism, the DIU-led governance, and the use of existing authorities were all genuine attempts to route around the constraints of traditional programs of record. Whether those workarounds are sufficient depends on whether industrial capacity or acquisition authority is the binding constraint. The evidence so far suggests it is both, and that the initiative’s next phase will require as much investment in domestic manufacturing throughput as in contracting velocity.

Sources