The IVAS Delusion
The US military has tried and failed to build a tactical helmet-mounted display for nearly 35 years. Can Anduril actually make it a reality?
Editor’s note: this article has been updated with comment from Anduril.
The US military has spent decades trying to deliver augmented reality to combat troops on the battlefield. Now, a Silicon Valley upstart may have finally pulled it off.
On Monday, defense unicorn Anduril unveiled its much-anticipated “EagleEye” augmented reality helmet-mounted display (HMD)1 at the Association of the US Army’s annual exposition in Washington, DC. Developed in collaboration with tech giant Meta and billed as a “modular, AI-powered family of systems that unifies command and control, digital vision, and survivability within a single, adaptive architecture,” EagleEye aims to provide soldiers with enhanced situational awareness by integrating live battlefield data, sensor feeds,2 and artificial intelligence-driven targeting overlays directly into their field of vision through a sleek hardware package reminiscent of Oakley glasses.
In early September, the Army awarded Anduril a $159 million contract for EagleEye as part of the Army’s Soldier Borne Mission Command (SBMC) program, the successor to the Integrated Visual Augmentation System (IVAS) effort that prime contractor Microsoft partnered with the company on in February following years of setbacks. But as an Anduril spokesperson told Laser Wars, the project had been in the works long before current program demand.
EagleEye is “an Anduril-funded internal R&D project that demonstrates the next step: a helmet-native, fully integrated version of our software,” the spokesperson said. “We’re continuing work on SBMC and SBMC-A,3 applying many of the lessons learned from those efforts. The key distinction is that EagleEye’s architecture is integrated within the helmet itself, rather than attached to it as in the SBMC configuration.”
Watch footage of Anduril’s EagleEye helmet-mounted display in action:
If this vision for EagleEye sounds familiar, that’s because it is. Like powered armor and jetpacks, HMDs have been a fixture of science fiction for decades,4 empowering fictional troops with real-time data and AI assistants to give them a tactical edge amid the chaos of combat. Indeed, virtual reality pioneer and Anduril cofounder Palmer Luckey frequently invokes sci-fi imagery to describe the company’s HMD efforts, comparing his vision for the Army’s future system to the immersive display inside the helmet of Tony Stark’s Iron Man armor, among other fictional battlesuits.
“Tactical heads-up-displays that turn warfighters into technomancers and pair us with weaponized robotics were one of the products in the original Anduril pitch deck for a reason,” Luckey wrote in a February blog post detailing the company’s new IVAS partnership with Microsoft. “Anduril [is] in a position to make this type of thing actually useful in the way military strategists and technologists have long dreamed of, ever since Robert Heinlein’s 1959 novel Starship Troopers.”5
Dreaming up a sci-fi-inspired combat system is one thing, but actually building it is another thing entirely. For 35 years, the Army has been chasing some version of this vision: a sophisticated yet rugged HMD that can deliver critical information to combat troops in the heat of battle. And again and again, the service has failed to bring it to life.
Soldier Integrated Protective Ensemble
The Army’s first serious attempt at a networked infantry display system began in 1990 with the Soldier Integrated Protective Ensemble (SIPE) Advanced Technology Demonstration, which imagined a fully-integrated combat suit including specialized headgear with integrated communications and thermal vision interfaces, advanced armor, and “an individual soldier computer that provided enhanced capabilities in navigation and message management,” as a 1992 Army Research Institute (ARI) analysis put it. The centerpiece of the system was an HMD meant to furnish soldiers with real-time data and environmental awareness through a visor-based display.
When demonstrated at Fort Benning, Georgia in 1992, certain elements of the system received “rave reviews,” according to a 2002 analysis from a group of Army engineers in the journal IEEE Pervasive Computing, particularly the ability to feed imagery from soldiers’ AN/PAS-13 Thermal Weapon Sights into helmet displays that allowed them to peek around corners or over foxholes without exposing themselves to enemy fire. The 1992 ARI assessment indicated that the “enhanced communications capability and thermal sight on the rifle, as well as some items of clothing, were deemed very acceptable” by soldiers tasked with putting SIPE through its paces.
“The Headgear Integrated Subsystem (HIS), Weapon Subsystem (WSS) and the Individual Soldier Computer (ISC) significantly enhanced lethality by allowing the soldier to detect, identify, acquire and engage enemy targets at increased ranges during both day and night and with improved accuracy,” according to a 2000 technical report on the program for the Army’s Natick Soldier Systems Center. “The HIS, WSS, and ISC proved to be vital to increasing the squad leader’s capability to communicate with both superiors and subordinates, as well as to exercise more positive command-and-control over personnel, weapons, equipment, information and procedures.”
While SIPE was a bold leap forward, it predated the microelectronics and materials science needed to make such a system practical in the field. With its analog components, unwieldy optics, and limited onboard computing adding more than 40 pounds to a soldier’s load, the system was more a burden than a boon. SIPE’s backpack-sized computer-radio-GPS weighed more than 18 pounds, while the HMD itself weighed nearly eight pounds and required a 15-pound power source to actually power the display, according to the Pervasive Computing analysis. And while the ARI assessment found that SIPE “demonstrated considerable potential for enhanced soldier capabilities and operational effectiveness” during testing, the system was far too heavy for soldiers facing a high-intensity firefight, so much so that “the helmet assembly was so heavy and unbalanced that soldiers were typically unable to lift their heads and fire from a prone position,” according to a 1996 Government Accountability Office report.
“Although the Army regarded SIPE as successful, for many test conditions, the SIPE-equipped soldiers failed to outperform soldiers using standard equipment,” the report says.
Beyond weight, system testing in 1992 yielded a plethora of complaints from soldiers on just how uncomfortable SIPE’s HMD was. “The problems with the visor caused many of the problems attributed to the HMD and the discomfort from the visor masked the potential value of the HMD,” as the ARI assessment put it. “Additionally, the soldiers complained that the HMD blinded them to the periphery and cut off their normal field of vision. It illuminated their faces at night, and also caused vertigo. One soldier cautioned, ‘never turn your head in SIPE — it makes you car sick because your eyes are seeing things that your brain isn’t expecting because your head isn’t moving.’”
Land Warrior
Such problems persisted with the Army’s Land Warrior program. Formally approved in 1993 and directly inspired by the capabilities showcased during the SIPE demonstration, Land Warrior was the Army’s first real effort to bring digital capabilities to the dismounted soldier. The system included an integrated helmet-mounted monocle display (known as the Integrated Helmet Assembly Subsystem), wearable computer, GPS, communications equipment, and a weapon-mounted camera that allowed soldiers to aim around corners.
Like SIPE before it, Land Warrior’s performance was decidedly insufficient to merit widespread fielding. A 2006 GAO report found that the system offered enhanced situational awareness, but was hindered by poor reliability and excessive weight of nearly 13 pounds that “did not demonstrate sufficient soldier acceptance” due to usability challenges in operational environments and frequent complaints of physiological discomfort from helmet-mounted components. A contemporaneous Popular Mechanics feature on system, which actually deployed to Iraq with the US Army’s 4th Battalion, 9th Infantry Regiment the same year as the program’s cancellation, bears this out:
There’s a half-billion dollars invested in the gear hanging off the heads, chests and backs of the soldiers of Alpha company. Digital maps displayed on helmet-mounted eyepieces show the position of all the men in the unit as they surround a block of concrete buildings and launch their attacks. Instead of relying on the hand signals and shouted orders that most infantrymen use, Alpha company communicates via advanced, encrypted radio transmissions with a range of up to a kilometer. It’s more information than any soldiers have ever had about their comrades and their surroundings.
But as Alpha kicks in doors, rounds up terror suspects and peals off automatic fire in deafening six-shot bursts, not one of the soldiers bothers to check his radio or look into the eyepiece to find his buddies on the electronic maps. “It’s just a bunch of stuff we don’t use, taking the place of useful stuff like guns,” says Sgt. James Young, who leads a team of four M-240 machine-gunners perched on a balcony during this training exercise at Fort Lewis, Wash. “It makes you a slower, heavier target.”
After investing more than 10 years and $500 million in the program, the Army terminated Land Warrior in fiscal year 2007, with service officials stating that the system “really had some problems” despite incremental updates intended to reduce the system’s weight and bolster survivability. While various program elements were rolled over into the subsequent Ground Soldier System effort, later renamed Nett Warrior, subsequent attempts at robust HMDs never seemed to fully catch on.6
Integrated Visual Augmentation System
More than a decade after Land Warrior’s demise, the Army opted to take another concerted run at a tactical HMD with IVAS, a ruggedized version of Microsoft’s HoloLens 2 mixed-reality display. As before, IVAS sought to outfit soldiers with a suite of sensors and, this time, an onboard artificial intelligence system to “provide the improved situational awareness, target engagement, and informed decision-making necessary to achieve overmatch against current and future adversaries,” as the service put it.
Yet early versions of IVAS encountered the same issues as its predecessors. The US Defense Department Director of Operational Testing & Evaluation’s (DOT&E) fiscal year 2021 assessment of the system revealed that soldiers who evaluated early versions of IVAS “lack[ed] confidence in their ability to complete the most essential warfighting functions effectively and safely while wearing the IVAS in all mission scenarios.” The primary reasons were physiological: the system’s design and weight (a comparatively light 3.4 pounds) forced soldiers to endure symptoms of cybersickness like headaches, nausea, and neck strain.
The following year’s DOT&E assessment of the system was equally brutal. An Army infantry company that tested updated versions of the system, known as IVAS 1.0, during an operational demonstration reported that it “was more successful accomplishing their operational missions with their current equipment” than with the high-tech goggles,
“The majority of soldiers reported at least one symptom of physical impairment to include disorientation, dizziness, eyestrain, headaches, motion sickness and nausea, neck strain and tunnel vision,” the DOT&E report continued. “Soldiers cited IVAS 1.0’s poor low-light performance, display quality, cumbersomeness, poor reliability, inability to distinguish friend from foe, difficulty shooting, physical impairments and limited peripheral vision as reasons for their dissatisfaction.”
Compounding these issues was a lack of clear criteria among Army leaders for determining whether rank-and-file soldiers would actually embrace IVAS as a useful battlefield asset. A 2022 Defense Department Inspector General audit of the system concluded that the service had not previously “[defined] suitable user acceptance measurements for testing and evaluation,” an oversight that could “result in wasting up to $21.88 billion in taxpayer funds to field a system that soldiers many not want to use, or use as intended.”
The Army tried to remedy these problems with a series of fixes in its next variant, IVAS 2.0, including a new form factor designed to address the physiological issues identified during testing, but the damage was done. By December 2023, the service was already soliciting the defense industry for a follow-on effort called “IVAS Next’” that, by March 2025, was renamed Soldier Borne Mission Command. The following month, the Army approved the transfer of Microsoft’s IVAS contract to Anduril. The service did not include any funds for IVAS procurement in its 2026 budget request released the following month. At the moment, it appears the most action IVAS will ever see is at the southern US border with Mexico, but that’s about it.
EagleEye and the Future of Military HMDs
The history of SIPE and Land Warrior demonstrates a simple truth: while capabilities alone may dazzle US military decision-makers, they don’t win adoption from service members — ergonomics, endurance, and trust do.
SIPE’s promising optics and thermal fusion were overwhelmed by unacceptable weight, visor problems and physiological effects that prevented soldiers from carrying out the most basic tasks, a usability failure that no amount of capability could paper over. Likewise, Land Warrior repeatedly ran into similar reliability, weight, and physiological issues, as did IVAS — challenges that crystallized this persistent gap between headset capabilities and soldier acceptance. These failures offer a technical lesson to the Army: a truly effective HMD must feel invisible during operations, not become a new burden or distraction.
That history suggests three blunt imperatives for EagleEye. First, the system must prove it does not impair core soldier tasks in controlled live-fire and maneuver exercises due to excess weight or persistent discomfort. Second, EagleEye must demonstrate endurance: realistic multi-hour mission profiles with metrics for power, thermal and mean time between system aborts that match real-world operational tempos. Third, human factors must be validated under stress in repeated field trials — after IVAS, the consistent absence of physiological symptoms are critical to winning over soldiers during testing touchpoints. If EagleEye meets those requirements, it will likely overcome the failures that sank its predecessors.
Does Anduril have a path forward? Potentially. Anduril’s stated approach — pairing consumer-grade display and waveguide tech (Meta/Qualcomm) with a defense-first, AI-enabled software integration and network layer (its Lattice software suite) — address important issues the Army lacked in prior iterations such as efficient compute, modular hardware and software, and a more streamlined form factor. Indeed, Luckey appears extremely confident in EagleEye’s ability to mitigate the past physiological issues in particular given his experience developing commercial augmented reality systems.
“I have got this shit figured out,” Luckey told reporters last week ahead of the system’s debut at AUSA, according to Breaking Defense “I’ve done this before. I’ve done it more or less perfectly.”7
But countering cybersickness isn’t all that matters: execution in realistic conditions, honest testing criteria, and a willingness to strip advanced features that hurt survivability will determine the success of the SBMC effort more than any single capability.
Anduril isn’t the only company gunning to build the SBMC. The Army also awarded $195 million contract to Palantir-backed startup Rivet Industries in September for its similarly glasses-style augmented reality system.
“Rivet’s philosophy is simple: technology should serve the soldier, not the other way around,” Rivet Industries CEO Dave Marra told Laser Wars in an statement. “We’re building systems that disappear into the mission — capabilities that enhance soldier lethality without demanding more of the operator. You can see their fingerprints in every detail because this system has been shaped by their feedback and refined by their experience.”
”We’re not chasing flash; we’re chasing trust, and trust isn’t permanent. It’s earned when the daylight is gone, the radios are hot, and fatigue sets in after miles hiked and rounds fired,” he added. “Soldier Borne Mission Command represents more than a contract; it’s a generational responsibility to redefine how the infantry fights and wins. That’s the standard we’ve set for ourselves, and we’re going to earn it every day through conviction, humility, and relentless focus on delivering what the mission demands.”
The pattern from SIPE to Land Warrior to IVAS is clear: new capabilities win only when they stop feeling new. EagleEye’s promise — better sensors, smarter software, and a relatively comfortable form factor — addresses real technical gaps, but the SBMC program will live or die on hard, soldier-facing metrics of live-fire performance, mission-length runtimes and repeatable results under stress. If Anduril can prove the system is effectively invisible to the operator when it must be yet still decisive when it’s needed,8 it will finally crack the classic military capability vs. burden tradeoff.
In the end, EagleEye won’t be judged by its sizzle reel or marketing hype — it will be judged by whether soldiers actually want to wear it when the shooting starts.
While the US Army frequently uses phrase “heads-up display” (HUD) interchangeably with “helmet-mounted display,” the former is inaccurate. A HUD projects information on a static surface like an aircraft canopy or vehicle windshield, while an HMD projects information onto a helmet visor so it moves with the user’s head.
The sensor nodes integrated into the top of the helmet are purportedly referred to as “wolf ears” at Anduril, although “cat ears” might be more appropriate.
Soldier Borne Mission Command-Architecture will be “the software backbone for the Army’s new mission systems,” as Anduril describes it.
Thought not technically an HMD, American inventor Albert Bacon Pratt patented the design for the very first helmet-integrated weapon system, known as the “gun helmet,” in 1916.
Here’s how Heinlein described his vision of an HMD in Starship Troopers: “Since your head is the one part of your body not involved in the pressure receptors controlling the suit’s muscles, you use your head — your jaw muscles, your chin, your neck — to switch things for you and thereby leave your hands free to fight. A chin plate handles all visual displays the way the jaw switch handles the audios. All displays are thrown on a mirror in front of your forehead from where the work is actually going on above and back of your head. All this helmet gear makes you look like a hydrocephalic gorilla but, with luck, the enemy won’t live long enough to be offended by your appearance, and it is a very convenient arrangement; you can flip through your several types of radar displays quicker than you can change channels to avoid a commercial — catch a range & bearing, locate your boss, check your flank men, whatever.”
To be fair, the Enhanced Night Vision Goggles-Binocular (ENVG-B) system the Army began fielding to soldiers in 2019 does incorporate some of the augmented reality features that the service envisioned with previous HMD efforts, including map overlays and integration with the Family of Weapon Sights-Individual.
A 2022 study in the journal Virtual Reality suggests that cybersickness and other physiological effects of helmet-mounted displays remain pervasive regardless of device.
Heinlein alluded to dynamic in his description of powered armor in Starship Troopers: “The real genius in the design is that you don’t have to control the suit; you just wear it, like your clothes, like skin. Any sort of ship you have to learn to pilot … but a suit you just wear.”