24JUL

Pentagon orders 120 drones in five weeks

2 min read

18:27UTC

The Pentagon advanced 19 of 49 firms in its Drone Dominance programme on 2 July, ordering each survivor to build 120 armed drones within roughly five weeks.

← Back to BAE unveils Britain's first combat drone Jump to analysis ↓

TechnologyDeveloping

Key takeaway

The Pentagon cut Gauntlet II to 19 firms and set a five-week test of who can actually build.

The Pentagon advanced 19 of 49 companies to the final stage of its Drone Dominance programme on 2 July and ordered each survivor to build and deliver 120 armed drones within roughly five weeks ¹. Drone Dominance is the US Department of Defense effort to mass-produce cheap attack drones; the 49 firms had tested at Camp Grayling, Michigan, through 20 June , and this cut drops 30 of them .

Delivering 120 armed drones in five weeks tests a factory, not a prototype. Stage one ranked prototypes on performance; this stage ranks production lines. A company can fly the best drone at Fort Carson, Colorado, in August and still be dropped for lacking a line that turns out 120 airframes on a war footing. That is the filter the programme, worth $1.1 billion for up to 60,000 one-way attack (OWA) drones, now applies.

The survivors include Teal Drones, the Red Cat subsidiary ; Neros, which cleared 43% acceptance in the earlier FPV round ; and Skycutter, tested alongside Ukrainian entrants in June , plus Auterion, ModalAI, XTEND and the Ukrainian firm General Cherry. No production awards follow until the August live-fire, so the field may thin again before the money moves.

Deep Analysis

In plain English

The Pentagon has been running a competition called Drone Dominance to find companies that can build cheap attack drones, the kind that fly into a target and explode, at massive scale. In the first round, 49 companies had their drones tested at Camp Grayling, a US Army base in Michigan. On 2 July, the Pentagon picked 19 winners and cut the other 30. But instead of just ordering drones from the 19, it told each one to build and deliver 120 armed drones within about five weeks, before an August test-firing at Fort Carson, an Army base in Colorado. This checks whether a company can actually manufacture drones at speed, rather than only design one good prototype.

Deep Analysis

Root Causes

RUSI's finding that only around 10% of drones in a 100-150 UAV salvo against a well-defended target typically get through is the operational logic behind demanding proof of manufacturing volume rather than flight performance alone: a mass-attrition doctrine only works if a production line can replace losses continuously, which a one-off prototype cannot demonstrate.

Phase 1's own shortfall matters more directly. Neros's 43% acceptance rate on contracted Archer units left the programme roughly 10,000 units short of its 30,000 target, and that specific gap, not a general doctrine shift, is why Gauntlet II tests delivery volume before committing further money.

What could happen next?

Consequence
A further culling of finalists is likely before the Pentagon commits full production money, since Gauntlet II adds a manufacturing test on top of the performance filter Stage 1 already applied.
Opportunity
Firms that clear the 120-drone build test gain a strong reference case for allied procurement offices assessing production readiness beyond flight performance, similar to how Skycutter's Phase 1 score opened UK and European conversations.

Source Landscape

This story draws on neutral-leaning sources

Primary parallel: The US Navy's A-12 Avenger programme passed its early design-review and performance milestones in the late 1980s, then collapsed in 1991 when contractors could not show they could build the stealth bomber within cost and schedule, and the Pentagon cancelled the whole programme rather than accept a production shortfall.

Counter-parallel: Ukraine's own FPV drone sector scaled from garage prototypes to hundreds of thousands of units a year after 2022 without passing through a formal government production-readiness gate at all, showing a wartime production base can scale under market pressure alone, without the phased test Gauntlet II now applies.

Consensus view: CNAS's Stacie Pettyjohn reads the shift from Stage 1's flight trials to a 120-drone, five-week build test as the Pentagon deliberately importing the lesson of its own Phase 1 shortfall, when Neros shipped all 2,400 contracted Archer drones but only 43% passed acceptance; this stage tests whether a firm can manufacture at rate before any production dollars move.

Counter-view: Analysts at Ukraine's Come Back Alive project argue five weeks is itself a generous bar. Ukrainian FPV workshops have run 48-hour build-to-field cycles under active bombardment since 2023, so a Pentagon test built around a five-week, peacetime cycle measures against a production tempo the war has already exceeded by a wide margin.

Key tension: Whether the Gauntlet II production test reflects genuine industrial-mobilisation readiness, or a slow-motion Pentagon version of a wartime benchmark Ukraine's own manufacturers cleared years earlier.

Sources:The Defense Post

Mentions:Pentagon →Drone Dominance →Camp Grayling →Fort Carson →Neros →Skycutter →General Cherry →Northrop Grumman →Golden Dome →Joint Laser Weapon System →Michigan →US Navy →Gauntlet II →Ukraine →US Army →

First Reported In

Update #14 · UK's £5bn drone bet follows Healey's exit

The Defense Post· 5 Jul 2026

Read original →

Causes and effects

Caused by

Drone Gauntlet now judges the parts

Stage 1 closed at Camp Grayling on 20 June, setting up the cut to 19 finalists.

Occurred 20 Jun 2026

Read story →

Drone Dominance Gauntlet opens 8 June

Stage 1 opened at Camp Grayling in early June, testing the 49 companies later cut to 19.

Occurred 8 Jun 2026

Read story →

Pentagon's drone buy lands a third short

Neros's 43% FPV Phase 1 acceptance rate carried it into the Gauntlet II final stage.

Occurred 7 Jun 2026

Read story →

Two Ukrainian firms enter Pentagon gauntlet

Skycutter's June Phase 2 entry alongside Ukrainian firms carried through to the final-stage cut.

Occurred 3 Jun 2026

Read story →

This Event

Pentagon orders 120 drones in five weeks

The five-week, 120-drone order turns Gauntlet II from a fly-off into a factory test, separating prototype shops from real production lines.

Led to

Gauntlet II sets five-week build sprint

The Fort Carson build sprint mechanics follow directly from the Pentagon's 2 July cut to 19 Gauntlet II finalists.

Occurred 3 Jul 2026

Read story →

Different Perspectives

Australian Department of Defence

The $6.9 million counter-drone award to DZYNE, part of Ondas's $70 million order book, is a funded delivery against a real budget line while most of Farnborough's reveals carry no disclosed contract value.

Baykar

Astore Levante extends the Bayraktar TB3 into a European joint venture the same way the Kizilelma export deal opened Indonesian production, treating proven Turkish airframes as the basis for allied manufacturing rather than new design programmes.

Leonardo

Declaring LBA Systems fully operational converts a Turkish export relationship into Italian production lines for Astore Levante, giving Italy a domestic combat-drone manufacturing base without funding a new airframe from scratch.

Minneapolis City Council (Vice President Jamal Osman)

The council killed a free Skydio trial 6-6 over the company's Israeli and ICE sales, not its drone-as-first-responder technology. It sets a precedent that other councils evaluating DFR vendors will now have to weigh alongside cost and capability.

UK Ministry of Defence

BAE's Brontanax reveal at Farnborough runs alongside the MoD's own Project NYX and Project Corvus competitions, neither of which has named Brontanax a winner. A demonstrator and a £300 million programme name are not yet a production order.

Procurement sceptics

Two identically sized IDIQs to different primes within seven weeks, and a five-nation pact where one partner moves weeks ahead of the rest, could just as easily read as an industrial base still improvising vendor mix as a deliberate hedging doctrine. Neither ceiling appears sized against a validated requirement yet.