An image of the gun and portions of the turret and chassis of the prototype M1E3 tank, which will be on display at the North American International Auto Show in Detroit on Wednesday. (U.S. Army)
When the North American International Auto Show opens in Detroit on Wednesday, more than 250,000 expected visitors will have a chance to see a rare and unique vehicle among the 500 SUVs, trucks and minivans.
It has seating for three, improved fuel mileage, better off-road handling and lots of high-tech bells and whistles.
It comes in one color — Army green.
Standard equipment includes a 120mm smooth-bore cannon.
“This is a U.S. Army display on the M1E3 Abrams,” said Ashley John, an Army spokeswoman for the military display at the car show, on Friday.
The Army hopes the M1E3 prototype will become its main battle tank for decades to come. It’s meant to be lighter, faster, more agile, secure and lethal than any armored opposition it faces on the battlefield.
The service is requesting $723.5 million in the 2026 National Defense Authorization Act to develop the tank. Roush, a Michigan-based design firm, produced the prototype. If the tank goes into production, it would be built at the General Dynamics Land Systems factory in Lima, Ohio.
The Detroit showcase comes at a time when tanks’ dominance on battlefields is in question for the first time in over a century.
“The fundamental idea has been that for your army to advance, you needed armor to break through — and to destroy your enemy’s armor,” said Michael Bohnert, an engineering analyst with the research firm Rand Corp.
Russia’s invasion of Ukraine has undercut the tanks’ traditional tactical advantage, with relatively inexpensive drones, loitering munitions, precision artillery and strikes by a new generation of highly precise rockets and missiles.
Russia has lost more than 4,300 tanks since the invasion began in February 2022, including top-end T-90 tanks, according to Oryx, an open-source defense research site.
A destroyed Russian tank in the Donbas region of Ukraine. The cage-like material on top and on the side of the turret is an improvised protection from top-down drone attacks, meant to cause explosives to go off before striking the tank itself. The photo is one of thousands compiled by Oryx, an open-source research tool chronicling destroyed tanks and vehicles in the war in Ukraine. (Oryx)
Ukraine has lost an estimated 800 tanks in the conflict, from Cold War-era T-72 and T-80 tanks it kept after the collapse of the Soviet Union to American M1A2 Abrams, British Challenger 2 and German Leopard 2 tanks it’s been supplied by allies since the war began.
In its Military Balance 2025 report, looking at conflicts around the world, the International Institute for Strategic Studies, a London think tank, called the tank losses the highest since World War II and “unsustainable” for both sides.
The high attrition rates have caught the attention of military leaders around the globe, including at the Pentagon.
U.S. Army Brig. Gen. Geoffrey Norman, director of the Next-Generation Combat Vehicle Cross-Functional Team, said in 2023 that the toll on tanks in Ukraine showed the U.S. had to “optimize” the Abrams if it was to remain “the apex predator on future battlefields.”
Until recently, optimization meant more. More armor, more optics, more external defensive systems. The M1A2 was slated for two rounds of system upgrades.
But all the additions also increased the weight of the Abrams, which ballooned to 78 tons. It made transport to the battlefield a logistics headache — a C-17 Globemaster III transport could only carry one tank at a time. The noisy, smoky diesel engine gave off a significant heat and noise “signature,” making it easier to spot on the battlefield. The weight could make maneuvering a slog in muddy or uneven terrain.
The Pentagon pivoted, and the upgrades to the M1A2 were trimmed or ditched.
Idaho National Guard soldiers train on the M1A2 Abrams at Fort Bliss, Texas, in August 2025. (U.S. Army)
An M1A2 Abrams tank fires its 120mm gun during training at Camp Ripley in Little Falls, Minn., July 29, 2025. (U.S. Army)
The new tank is not a “clean sheet” model, designed from the treads up. But the “E” in the designation M1E3 stands for “engineering,” signaling that the new tanks are a significant departure from earlier models.
The few renderings of the new tank appear to show the Army sticking with the 120mm smoothbore M256 cannon. However, Army planners have sought the ability to use self-propelled hypersonic rounds and anti-tank guided munitions in place of standard armor-piercing and anti-personnel rounds.
A Congressional Research Service report said the Army was expected to incorporate many of the suggestions from a 2019 U.S. Army Science Board report on specifications for a “fifth-generation” tank.
Improvements included reducing the crew from 4 to 3 by eliminating the loader and adopting an automatic ammunition system. The tank would have a hybrid electric drive, which would enhance performance, range, and reduce the noise and heat “signatures” of a diesel-only engine. An artificial intelligence system aided with parsing images from the battlefield to better detect threats and rank targeting priorities for the gunner.
Bohnert, the Rand analyst, said drone technology and tactics are rapidly changing. Tank designers must find ways to destroy, jam, or confuse enemy munitions to retain a leading role in combat.
Initially, the developers of the M1E3 tank wanted to field it by 2030. But Army officials pushed for deliveries in 2026, even as changes were made during the process.
Gen. Randy George, the Army chief of staff, said earlier this month that four prototypes will be delivered to undisclosed Army bases this year.
What the M1E3 does in combat won’t be known until well down the road. The prototype on display in Detroit, which will remain stationary, will show off the exterior — no peeking inside.
The Army says the M1E3 is a work in progress. Bohnert says that is all any tank can be today.
“Everyone is trying out their own vision of a solution,” Bohnert said. “You can’t afford to wait and see. You have to keep on experimenting. It is going to take a lot of failure-based learning, which is not great if a human life is involved. Whoever figures it out is going to make a lot of money and save a lot of lives.”
