Welcome to Words with Wynn! If this is your first time perusing my content, and you’d like more of my weekly musings, subscribe below:
Last month, Anduril and Meta announced a partnership to help push the boundaries of augmented reality (XR) for the U.S. military. The details are fairly high level at this point, but the gist is that Anduril will be the prime with Meta working as a technology provider. It sounds like Anduril will be responsible for manufacturing the hardware itself (helmets, goggles, HUDs etc.) and integrating the tech stack (Eagle Eye) with its broader Lattice network as well as other DoD technologies, while Meta will provide its XR expertise, IP, and AI models. This is an incredibly exciting ‘getting the band back together’ moment for Luckey and Zuckerberg to reunite years after Palmer’s unceremonious removal from Meta in 2017. In his Core Memory interview (linked below), Palmer has said that this collaboration allows him to re-access much of what he developed internally while building Oculus then Meta, and supercharge it with their work at Anduril. The announcement builds on the momentum of Anduril’s previous XR coup back in February: taking the prime position on the DoD’s $22B Integrated Visual Augmentation System (IVAS) program from Microsoft.
This got me jazzed, not only because it portends a future of technomancer American supersoldiers..
…but because I peeled the onion back a bit, and went deep on Palmer Luckey’s interview with Ashlee Vance. That discussion gets into the meat and potatoes of what this partnership means and what Palmer really has in mind for the future of reality-enhanced American warfighters.
In that meandering hour, there was a specific minute that really made my ears perk up:
“So, the way that I look at something like Eagle Eye is that it should not be one display system that does everything for everybody. It’s actually an architecture that allows- like, EagleEye can basically, it can host many different types of vision augmentation systems. It can host head mounted displays from multiple different companies, including companies that don’t even work directly with Anduril. [We’re] building an open architecture where the government owns the interconnects. They own the standards and they can work with other companies to build different displays that plug in.
The reason for that is that I don’t think we can build one display that does everything for everybody, right? The guy who is a logistician needs a different thing than the guy who's a helicopter pilot who needs something different than the guy who's an infantryman breaking down doors in the middle of the night. And in terms of the units, which is what you're asking about, I think that within the decade, you're going to see a head-mounted display of some kind on basically every single head in the DoD.
They're going to use it for different things. Some guys are just going to be checking their emails on a secure terminal. Others are going to be shooting people in the face. And you need to be able to support all those use cases. I don't see this as just like 10,000 units or 20,000 units. I see this as outfitting the entire armed forces. And so everything I'm doing is to try and build an architecture that scales to that.
…
A guy who's like, let's say, an army aviation maintainer who's working on helicopters, he probably needs something that is like a lightweight pair of Oakleys that is able to display work instructions, remote help, labeling particular things, checklists. He does not need an ultra-wide field of view, thermal, UV, hyperspectral vision augmentation system like a frontline infantryman might. And so you kind of have to build something that unifies those two things so they can share one set of software, one set of apps, one set of common standards.” - Palmer Luckey on Core Memory
Anduril isn’t building a product, it’s launching a platform. And this framing sounded awfully familiar to me:
“I'm pretty sure glasses and holographic presence and AR is going to be a completely ubiquitous product, right, it's just like everyone who had a phone before replaced it with a smartphone and then a lot of more people got smartphones. If all we get is all the people in the world who already have glasses upgrading to glasses that have AI in them, then this is already going to be one of the most successful products in the history of the world, and I think it's going to go a lot further than that. So I know there's that.
There is the thing about controlling our own destiny. It's strategically valuable. You know we did this calculation or estimate at some point where it's like how much money do we lose from our core family of apps to the various taxes that the platforms [Apple’s App Store, Google, etc.] have to like, when they tell us we can't run the ad business the way that we think we should be able to.
When they tell us we can't ship certain products, that way people use the things less or like them less, and it's hard to exactly estimate it. But I think we might be twice as profitable if we own the platform or something. I think from that perspective that's worth a lot, just from a pure dollars perspective, which is not primarily how I come at this stuff…” - Mark Zuckerberg on Acquired
The latter Zuckerberg quote was the inspiration for another Words with Wynn, Pieces on the Board: The Power of Platforms, where I basically pointed out that emerging AI startups and legacy corporates (see: OpenAI investing in Figure, Google investing in Apptronik) were all too happy to pour mountains of treasure into the coffers of humanoid robotics companies for a call option on a platform that could dwarf anything we’ve ever seen. Should that become the case, no one wants to be beholden to a toll-extracting App Store equivalent.
“Owning your own destiny [platform] has proven time and again to be fundamental in cementing network effects with hardware and user lock-in within your ecosystem. It’s tech strategy 101.” - Me, Pieces on the Board
Thus has been the logic with Zuckerberg’s massive VR investments as well.
Spelled out by the man himself, Meta’s unbelievable R&D bill potentially makes sense if an augmented reality piece of consumer hardware stands to be the next iPhone:
“AR/VR was all the hype around the time that Zuckerberg rebranded Facebook to Meta, and Wall Street torched him in the ensuing years for the company’s massive investments into augmented reality technology as the sector fell out of favor. But when you frame it within the context of owning the next platform, their hefty R&D bill starts to make a bit more sense. You can basically run a loose NPV for what that’s worth against the cash sink to develop the technology today. This theoretical number could be so huge that it’s rational to continue your investment as an option on a gargantuan future market. Whether or not AR/VR becomes the dominant platform is anyone’s guess, but if you’re the founder and controlling shareholder of Meta, and your timelines are measured in decades, there is a serious logic in betting on (and willing into existence) augmented reality if the probability of its ubiquity is high enough.” - Me, Pieces on the Board
Now, perhaps this is a case where the man with a hammer sees everything as a nail, but I don’t think I’m exaggerating when I highlight the underappreciated importance of this alliance in supercharging Anduril’s dominance as a future prime.
The partnership sounds benevolent when Palmer phrases it as the government owning the interconnects with an open architecture for other companies to build into the Eagle Eye ecosystem. But inherent to that dynamic is building atop Anduril’s platform, with value accruing accordingly. Even if it’s not a direct toll-taking relationship like the App Store, building the substrate for the military’s entire XR stack (third party or otherwise) to integrate into is an enviable strategic position to occupy.
It’s all platform. It’s all ecosystem. And if Palmer Luckey has his beautiful way, it will all be computer. Delicious, delicious Lattice-enabled computer.
This collaboration has the potential to be a win-win-win: leveling up the American warfighter, irreplaceably cementing Anduril as the DoD’s XR dealer of choice, and giving Meta another avenue to capitalize their ungodly R&D spend. Plus, if history is any guide, we’ve seen tools of war push the cutting edge for technologies (GPS, semiconductors, etc.) that eventually bleed into the consumer market, which still represents a significantly larger TAM than military-grade headsets. Meta recognizes this, and Luckey discussed this interchange directly in his AWE interview.
That said, there is also room for the reverse:
“It’s the Fairchild Semiconductor angle that is most interesting, however, specifically Bob Noyce’s resistence to getting too entangled with the U.S. government. His opposition was not ideological, but pragmatic. From Chris Miller in Chip War:
‘'The computers that guided the Apollo spacecraft and the Minuteman II missile provided the initial liftoff for America’s integrated circuit industry. By the mid-1960s, the U.S. military was deploying chips in weaponry of all types, from satellites to sonar, torpedoes to telemetry systems. Bob Noyce knew that military and space programs were crucial for Fairchild’s early success, admitting in 1965 that military and space applications would use “over 95% of the circuits produced this year.” But he always envisioned an even larger civilian market for his chips, though in the early 1960s no such market existed. He would have to create it, which meant keeping the military at arm’s length so that he — not the Pentagon — set Fairchild’s R&D priorities. Noyce declined most military research contracts, estimating that Fairchild never relied on the Defense Department for more than 4 percent of its R&D budget. “There are very few research directors anywhere in the world who are really adequate to the job” of assessing Fairchild’s work, Noyce explained confidently, “and they are not often career officers in the Army.’
Noyce’s gambit — most fully realized at Intel, the company he started after leaving Fairchild Semiconductor, in part because a venture-backed company offered far more upside — obviously paid off. The consumer market had massively more volume than the military, which enabled far larger R&D and capex budgets, which drove improvements in chips far more quickly than would have happened under the government’s direction. And, in the end, the Pentagon benefited: what undergirded the defense department’s successful modernization in the 1980’s was the wholesale adoption of computerized weapons systems that were based on technology that was built for the consumer market. - Ben Thompson, Anduril + Meta, Silicon Valley and the Pentagon, Meta’s Motivations
Where early government R&D helped to kickstart the U.S. semiconductor industry, it was ultimately the massive consumer market that provided the demand that powered the industry to track Moore’s Law. History doesn’t repeat itself, but it can rhyme. Meta can benefit from governmental use cases which will push the bleeding edge, while the budget to fund this kind of technology may ultimately be sourced from massive consumer demand.
A key aside – virtually all of Meta’s cutting-edge R&D up to this point has been financed off of its own balance sheet, and Palmer has explicitly said that Anduril will be moving forward with Eagle Eye irrespective of winning the $100M Army contract on which the companies jointly bid (per WSJ).
But what’s this mean for investors?
The inevitable lemming rush of VC’s leaping to back VR/XR projects with a DoD veneer need to recognize that, more likely than not, they will be funding apps, features, and add-ons for the mothership’s platform. Not that some major winners can’t emerge – plenty of billion dollar businesses were built on Apple’s App Store – but that they will be backing second movers that are beholden to a platform beholden to the government. Pricing will matter and value accrual must be considered.
And history would suggest as much. Anduril’s previous acquisitions have been one-off drone or weapons systems that could be integrated into its software-enabled hardware ecosystem; low-margin weapons platforms powered by high-margin Lattice. These individual roll-ups didn’t accrue value or command multiples like that of Anduril’s integrated platform:
“There are a lot of really interesting defense capabilities that are out there, particularly in hardware, that have kind of advanced through the prototyping phase– they’re capabilities that have been funded with research dollars, they’re platforms that are incredibly interesting and potentially valuable. They’re not necessarily highly integrated on the software, autonomy, artificial intelligence side of things, and make a lot of sense for us to go in and say ‘If we were to go in and make this part of the Anduril suite of capabilities, it would be more valuable than standing alone…’
These are not traditionally venture backed companies… And you know, we’re not buying them at Venture Capital software multiples, we’re buying them at Defense multiples… Now, the risk here is that if you see what Anduril’s doing– I think we’ve done eight or nine acquisitions to date– if you see what we’ve done in this space it would be easy to say like ‘Oh look, there all these very cool exit opportunities if you were to start a defense company.’ I just don’t think that’s true. If you’re a defense company out there raising at 20, 50, 100x on in-year revenues, like, we’re not a buyer of those companies.” - Trae Stephens on The Generalist
Investors should also consider the extent to which these types of VR/XR Eagle Eye add-ons will have broader consumer applications. The Anduril x Meta collaboration is for modular helmets/glasses manufactured by Anduril with open architectures for other vendors, not improvements and directly retrofittable upgrades for Meta’s Quest line.
Additionally, we could very well see augmented reality play out akin to modern computing. Original mainframe type computers were adopted by enterprises that had business use cases and could rationalize the high price tag against the efficiencies that would be gained across their organization. This familiarized a generation with computers, which then came down the cost curve simultaneously to now-tech savvy users seeing the benefits of this hardware for their everyday life. In the case of XR, the initial breakthrough may end up being military uses where immersive data visualizations (troop movements, autonomous system coordination, intel, etc.) are a superior form factor to current computer + monitor setups, with advancements in this domain feeding back out to consumer use cases like gaming or enterprise work. Of course, Meta has already pushed relentlessly to drive down the price tag on its Quests, so (finally) unlocking a ‘killer use case’ for VR should climb an accelerated adoption curve vs. the emergence of desktops.
The second interesting takeaway was Palmer’s discussion of margins and the evolution of military helmets themselves:
“And one of the things that Anduril’s doing is actually taking extremely low margin and making a very concerted effort to drive the price of the system down because… the army has, call it, $20B to buy headsets. If I price them at $100,000 each, then that means that there’s some number of headsets they can buy. If I price them at a tenth of that, they can buy 10x as many headsets. Which world do you think Palmer Luckey wants to exist? I want this on everybody.
So, I am doing my damnest to make this as low cost as possible, way beyond probably what is rational. I’m confident I could extract more money out of the DoD by playing the game, riding the margins as high as I can, taking advantage of the fact that what we have is so far ahead of every other competitor. I do not have it in my heart to do that.” - Palmer Luckey on Core Memory
I believe this is genuine and laudable. I too want our men and women in uniform to have world class technology at their disposal, and all the better if it can be done at a price that inflates our deficit marginally less. Then again, it isn’t (and was never going to be) the helmets that will drive Anduril’s profitability. It’s the software stack to power them. Any CEO should be happy to place hardware for a modest profit if they can capture high margin SaaS fees in perpetuity on the back-end, and even better if the switching costs are astronomical (once integrated) and your customer turns like a titanic.
In reimagining our army’s headware, he also made an interesting point that struck me about the various component costs:
“You can build some really interesting things that are far beyond what the United States has done so far on ballistic helmets. And I’d say, one of the reasons the U.S. hasn’t done these types of technologies in the past- It’s not because I’m this genius that’s smarter than everybody. It’s because it didn’t make sense when the helmet was this thing that’s supposed to cost a few hundred dollars. When a helmet was supposed to be this cheap thing, you put it on your head, it needs to work, but the whole system is pretty cheap. If you’re instead building this kind of very expensive thing where you’ve got sensors, computers, radios, displays, optics, speakers, microphones- all this stuff integrated. You now need to look at it as an integrated product. And when you look at it as an integrated product, you always want it to be lighter, right? So you start to ask, where is the best savings per dollar? The best weight savings per dollar.
If I wanted to pull a hundred grams out of this helmet, how can I do that the cheapest? Well, I’ll tell you what. It’s not pulling a hundred grams out of the electronics. It’s not pulling a hundred grams out of the optics. It’s not pulling a hundred grams out of the microphones. It turns out that the place to invest money to save weight is in the actual physical ballistic shell itself.
So, all the sudden it actually makes sense, in the context of an integrated system that, ya know, costs quite a bit, to say ‘you know what, my helmet skin is going to cost $2,000 and in doing so it’s going to weigh 2lb less than it would have weighed otherwise.’ And it starts to make sense.” - Palmer Luckey on Core Memory
This evolution could transform a soldier’s helmet from an elementary shell into a more ‘exquisite’ weapons system, as it’s outfitted with increasingly sophisticated widgets depending on the job. Loading down the helmet component of a soldier’s kit with increasingly sophisticated hardware begins to change the cost equation as you explore the possible steps to reduce weight distribution across the board. Given that the marginal cost to further miniaturize already cutting-edge electronics is massive, you actually turn to innovation in the shell itself, the largest weight component. But what this made me consider as well was the profitability profile for an exquisite helmet system vs. a singular shell with some modular tech attachments (night vision goggles, radio on the side, etc.). Is this a more profitable piece of equipment to manufacture if you’re capturing a blended margin with integrated electronics etc. instead of plain kevlar buckets?
It also begs the question: To what extent does a more complicated, integrated system fall prey to the ‘totaled car’ phenomenon of recent years where insurers more readily write off modern cars as totaled because the complexity and cost to fix vehicles loaded down with sensors etc. has made it more cost prohibitive to repair them?
I’m sure that bigger brains than my own are going deep on these considerations, and I’d imagine that many of the tech components that will be integrated into the helmet will A) be modular in nature and B) will displace cost/tech that would have been borne by a soldier in other parts of their kit anyway (backpack, belt, etc.). So, I’m sure it’s a complex calculus.
Ultimately, I think that this is an incredibly exciting partnership. The prospect of Silicon Valley leaning into supporting American deterrence (or whatever euphemism you want to use for American military might) in a time of escalating global tensions makes me optimistic for the future of our national security. And that’s a good thing. The bloody detail of business strategy within the less-than-free market dynamics of navigating the military industrial complex will continue to evolve, and spawn a generation of fascinating case studies.
The tailwinds for startups building in defense tech continue to mount as conflicts like Ukraine and Israel-Palestine have been rude awakenings for Western militaries. Government budgets have reacted accordingly as the imperative to modernize has become self-evident; the agility that Washington is attempting to reinject into the military contracting apparatus is just one of many examples. I have high conviction that first movers like Anduril and Saronic will be able to continue compounding their advantages (integrated ecosystems, economies of scale, etc.) for outsized returns, but the size of these markets and urgency of demand will certainly contain opportunities for ambitious builders and savvy investors.
- 🍋