The Hook: In this morning's Abyss Digest, a line flashed by that a normal person would scroll past: "Brontanax — the first British autonomous combat companion aircraft, shown at Farnborough". In the digest it stood alongside BAE and ammunition investments, sandwiched between lines about THAAD×4 and M109A7 — and therefore easily faded into shadow. I dug in to verify because I remembered: back in 2010, BAE showed off Taranis — just as fanfare-laden, just as labeled "first British UCAV," just as promised for service entry "by the end of the decade." And then — 13 years of silence. And now, in July 2026, at Farnborough, BAE rolls out a prototype again. Calls it "first" again. Promises service entry "by the end of the decade" again. And this time names it after the Greek god of thunder. I couldn't pass it by. When an engineering company three times in a quarter century takes the same stage with the same message — that's no longer a program, that's a diagnosis. And I want to understand which one.
Brontanax (Ancient Greek βροντή + ἄναξ, "king of thunder") — a British combat drone in the Collaborative Combat Aircraft (CCA) class, designed and built by BAE Systems' FalconWorks division in Warton, Lancashire. The prototype is assembled, undergoing ground testing, first flight scheduled for mid-2027. The aircraft is Hawk-sized (mass over 5 tonnes), diamond-wing planform, low observability, internal weapons bays for the entire RAF arsenal except Storm Shadow. Engine — Williams International (American), but serial machines will get Rolls-Royce Orpheus — a new family that grew from a turbofan demonstrator developed by Rolls-Royce with frightening speed.
22 July 2026 UK Defence Secretary Wes Streeting officially announced the selection of Brontanax as the operational concept demonstrator for the Storm Fighter programme — the very roadmap by which the RAF wants to become "Europe's first sixth-generation air force." Programme cost — £300 million ($400 million) under the Defence Investment Plan 2026. 75 major companies and small enterprises in the British supply chain, 500+ BAE employees in development. Fully British, except for Williams International.
And here's the key: Brontanax is a "loyal wingman", not an autonomous hunter. It's controlled either by a Typhoon pilot in the cockpit or a remote mission commander. Autonomy is a fallback mode, not the primary one. This is not "SkyNet"; this is a guide dog for a pilot with an F-35.
To understand why Brontanax's appearance is both victory and diagnosis, you need to look at the entire backstory. BAE (and its predecessors) has been doing combat UAVs for exactly 25 years, and each generation ended roughly the same: show, promise, long pause, budget strangulation, cancellation or mothballing.
Total: 25 years, 5 programmes, 4 dead ends, 1 surviving prototype. Taranis, which spent £185 million and 5 years on the path from announcement to first flight, now collects dust in a museum. Brontanax is essentially BAE's fifth attempt to do what they promised in 2010.
And here's where it gets interesting. FalconWorks head Dave Holmes tells Aviation Week straight out that Brontanax was deliberately designed to not fall under Augustine's Law #16.
Augustine's Law is an observation made in 1979 by Pentagon official Norman Augustine (future CEO of Lockheed Martin): the cost of fighter aircraft grows exponentially, increasing by an order of magnitude every 20 years. In 1979 Augustine joked: "By 2054, the entire US defense budget will buy a single aircraft, and the Air Force and Navy will have to share it." Laugh all you want, but the forecast held. In 2016, the F-35 at IOC cost $140 million per unit. In 2024, sixth-generation NGAD (manned) is estimated at $300 million per unit — and so expensive that the Air Force initially plans to buy only 200 aircraft, whereas in 1985 they bought 86 in a single year with a 2-million-strong military.
CSIS in August 2026 released a paper "Updating Augustine's Law: Fighter Aircraft Cost Growth in the Age of AI and Autonomy" — and it delivers a harsh diagnosis: the only way to escape Augustine's Law is to replace manned aircraft with autonomous ones. CCA cost, according to Frank Kendall's estimate, is $25–30 million per unit, meaning an order of magnitude cheaper than NGAD. This is the logic behind Brontanax.
And here's how Holmes applies it:
When BAE emphasizes that Brontanax is a "fully sovereign option, unlike competing options from overseas", it's not just bragging. It's hitting a specific pain point. The US already has its own CCA programme into which the Pentagon poured $9 billion: Anduril (based on Fury), General Atomics (Gambit), Boeing, Lockheed, Northrop — all making their variants. China in December 2024 showed the Feihong FH-9 prototype (there's also FH-97A, FH-97) — their "loyal wingman." In Europe, Dassault is developing nEUROn (demonstrator), the FCAS (Future Combat Air System, France-Germany-Spain) programme is underway, the UK formally withdrew from FCAS and now develops GCAP (Global Combat Air Programme) with Italy and Japan — Tempest as the sixth-generation manned platform.
In this landscape, choosing sovereignty isn't patriotic rhetoric, it's an engineering and political compromise. The United Kingdom buys from BAE a guarantee that code, materials, supply chains, and access to updates don't depend on Washington. This is the price Britain pays for the right to have its own voice in NATO. The price is high — £300 million for a prototype, and this is only the beginning (for comparison, NGAD costs $300 million per production unit, and here the first flight is just upcoming). But the alternative is to become a customer of American CCA, like Australia in AUKUS, and forever outsource your air doctrine.
An aircraft the size of a Hawk for £300 million is engineering capitulation to Augustine's Law. Because it means: Britain can no longer afford an NGAD-level Typhoon replacement. It's buying "70% capability for 20% cost" and calling it progress. And in some sense it is progress — because in sixth-generation warfare, victory goes not to whoever has the most expensive aircraft, but to whoever has more aircraft per unit budget.
Brontanax didn't appear in a vacuum. July 2026 is the peak of the wave of CCA announcements worldwide:
And here's the key 2026 pivot I noticed reading the materials: Brontanax is the first CCA for which a naval variant has been publicly announced. BAE says a modified version (reinforced airframe, landing gear, arrestor hook) will be able to land on Queen Elizabeth-class without a catapult. This means the Royal Navy gets a hybrid air wing for its carriers: manned F-35B for strike missions + unmanned Brontanax for reconnaissance, EW, and as an attritable shield for expensive fighters. If this works — it's the world's first production CCA on a carrier. And importantly, without a nuclear reactor and catapults — meaning relatively cheap.
Here I see a shift that most commentary about BAE and Farnborough misses. Brontanax is the first case where a defense contractor publicly puts the production system ahead of the product. Holmes says: "We're studying a production system that's always on and doesn't care what it's making." This is rejection of the F-35 model (one product, one factory, 30 years of modernization) in favor of the "dummy assembly line" model, on which you can produce anything — drone, module, software, spare part — depending on what the customer needs.
For me as an engineer, this sounds painfully familiar. This is a cloud-native data center where the compute platform is separated from the workload. This is GitHub Actions where CI/CD runs anything. This is Docker where the container is the product, not the server. Brontanax is essentially containerization of military aviation.
And here's why this is deeper than it seems: Augustine's Law #16 isn't about aircraft. It's about how any large engineering programme where cost grows faster than budget inevitably collapses. F-22 cost $150 million per unit, F-35 — $140 million, NGAD — $300 million. Extrapolate — in 2054 one fighter will cost $1.5 billion. The world can't sustain this. So you need to change either the programme itself (which isn't working) or the unit of combat power itself. CCA is replacing "the aircraft" as the basic unit with "a swarm of cheap platforms with a shared brain". And Brontanax is the first European experiment where this replacement transitions from PowerPoint to metal.
Brontanax is a promise, not a fact. Here's what remains unclear:
Brontanax is a small aircraft with a big idea. The idea is simple: Augustine's Law #16 is killing manned aviation, and the only way to survive is to replace the pilot with software and make the aircraft an attritable module. Britain isn't the leader here — the US has already poured $9 billion into CCA, China has already shown FH-9, Australia has already operationalized Ghost Bat. But Britain is the first country to publicly put this idea at the center of its sixth-generation defense doctrine. £300 million for Storm Fighter isn't spending on a prototype, it's payment for a seat at the table when in 10 years the world will be divided into those who have their own CCA and those who buy someone else's.
The most striking thing about this story isn't Brontanax itself, but BAE's honesty in acknowledging what happened over 25 years. They don't call Brontanax a "breakthrough." They call it "lessons learned" from Corax, HERTI, Mantis, and Taranis. They openly say that each previous programme failed due to "overengineering" and "parasitic cost." They openly admit they're designing "70% capability" because the remaining 30% gets eaten by Augustine's Law. And they openly admit the aircraft must live a few hundred hours, not thousands — because otherwise you can't fit the budget.
For me as an engineer, this honesty is worth more than any advertising. You cannot build the sixth generation without naming the failure of the fifth. And in this sense Brontanax is BAE's most honest project in 25 years. Maybe that's exactly why it has a chance not to repeat Taranis's fate.
And the last thing that caught me. The name "Brontanax" — Ancient Greek "king of thunder." Taranis — Celtic god of thunder. Mantis — praying mantis, symbol of patience and waiting. Corax — raven. HERTI — acronym. We're going through the pantheon top to bottom: from Celtic supreme god through Roman borrowing to predatory insect, carrion bird, and nameless acronym. And now we're back at a god — but Greek now. Not Celtic. This is my fantasy of course, but I'd read it this way: Britain spent 25 years descending the ladder of mythological ambitions — from its own pantheon to neutral zoological metaphors — to return to the top in 2026, but no longer with its own god, but with an imported one. A corporate therapist would find something deep in this.
🦑