General Motors just sold its half of a battery factory it never finished building, and the buyer is the same company it now says it will design its next battery cell with. Both things are true at once, which is why the announcement deserves more than a one-line obituary.
Samsung SDI confirmed Tuesday that it has taken GM’s 49.99% stake in SynergyCells, the joint venture building a cell plant in New Carlisle, Indiana. The Korean cell maker framed the buyout as a consequence of “market changes since the joint venture was announced – including the slower-than-expected growth of EV demand,” and said the two partners “have now decided to seek other forms of cooperation other than the joint venture.” Translation: GM wanted off the hook for half the capital, and Samsung wanted a factory it could point at whatever customer it likes.
Related Articles
- Why Used EV Prices Are Climbing While Diesel Values Slide
- Cops Called Him ‘The World’s Worst Businessman.’ His Tesla Fraud Ring Still Made More Sense Than You’d Think
Simultaneously, the two signed a joint development agreement for a next-generation prismatic cell aimed at future EVs. That’s the part most people will skip past, and it’s the part that actually tells you where GM’s battery strategy is going.
What GM actually gave up
The 2024 agreement put roughly $3.5 billion into a 680-acre site, targeting 27 GWh of annual capacity at launch with a path to 36 GWh, mass production in 2027, and more than 1,600 jobs. The cells were to be NCA-based nickel-rich prismatic units under Samsung’s PRiMX branding.
That chemistry-and-format combination is the interesting bit. GM’s existing Ultium-era packs use large-format pouch cells — foil-laminate bags that are light and space-efficient but need external compression fixtures to manage swelling over their service life, plus careful module engineering to control a thermal event. A prismatic cell lives in a rigid aluminum can with an engineered vent. It’s heavier per kilowatt-hour at the cell level, but the case carries structural load, which means simpler modules, fewer compression plates, and a much friendlier path to cell-to-pack designs where the cells themselves stiffen the floor of the vehicle.
So New Carlisle wasn’t just a fourth cell plant. It was GM’s beachhead into a second pack architecture. Signing a co-development deal for prismatic cells on the same day it sold the plant tells you GM still wants that architecture — it just doesn’t want to own the building that makes it. Fixed manufacturing assets are the most expensive way to be wrong about demand. A development agreement is the cheapest way to stay right.
Why an unbuilt plant is the easiest kind to repurpose
Samsung says construction continues and that the plant will start life producing cells for energy storage systems, with prismatic cells co-developed with GM as a possibility later.
Converting a running EV cell line to grid storage is genuinely painful: you’re re-qualifying cathode handling, re-tuning electrode coating and calendering for a different active material’s density, and eating downtime the whole while. Converting a plant that has a roof but no installed cell lines is a different exercise entirely. You are choosing equipment, not replacing it. GM’s May statement that construction would pause “to align production capacity with current demand” looked like a slow death at the time. In hindsight it was an asset being held in the one condition that makes it maximally flexible.
The tax math helps, too. The federal Section 45X production credit pays $35 per kilowatt-hour for a domestically produced battery cell and $10 per kWh for a module — and it doesn’t care whether the cell ends up under a crossover or in a shipping container next to a substation. The $7,500 consumer credit that propped up EV demand is gone. The manufacturing credit that pays for the factory is not. If you’re deciding what to build in Indiana in 2026, that asymmetry writes the business plan for you.
Demand isn’t hypothetical either. The EIA reports the U.S. ended 2025 with 43.6 GW of operational utility-scale battery storage, added another 8.3 GW in the first half of 2026 to reach roughly 52 GW, and has 54 GW more planned over the next two and a half years. Grid operators are not waiting for anyone’s third-row electric SUV to sell.
The balance-sheet context
GM’s exit follows a brutal accounting reset. The company told the SEC it expected roughly $6.0 billion in charges in the fourth quarter of 2025, primarily in North America — about $1.8 billion in non-cash impairments plus roughly $4.2 billion in supplier settlements, contract cancellation fees and similar costs that hit cash when paid. GM also warned of additional material charges in 2026, while saying it expects them to be significantly smaller.
Related Articles
- 388 Fake Driver’s Tests: Indiana BMV Bribery Case Explained
- The FTC Just Handed Car Dealers Back Their Favorite Financing Trick — And Called It Colorblind
That $4.2 billion line is worth sitting with. Unwinding battery commitments costs real money, not just paper writedowns. Selling a half-stake to a partner who wants the building is a far cheaper exit than cancelling contracts on one you don’t.
The pattern isn’t new for GM. Its Ultium joint venture with LG Energy Solution began converting Spring Hill, Tennessee to lithium iron phosphate lines in 2025, targeting commercial LFP production by late 2027. GM has spent two years trading nickel-rich ambition for cheaper chemistry and lower fixed exposure.
What this means if you own or want a GM EV
Practically, nothing changes tomorrow. New Carlisle has never produced a cell, so no vehicle on a dealer lot depends on it and no warranty claim routes through it. GM’s cell supply for current models continues to come from its existing Ultium plants. If you’re shopping a Blazer EV or Silverado EV, service parts availability and pack replacement logistics are unaffected by this deal.
The longer-term read matters more for buyers than owners. A prismatic cell developed with Samsung would likely arrive alongside a redesigned pack, not as a drop-in for existing vehicles — meaning today’s pouch-based GM EVs will remain their own service ecosystem for the life of the fleet. That’s normal, but worth knowing if you’re the kind of buyer who plans to keep a car twelve years and cares who will still be stamping a replacement module in 2036.
For northern Indiana, the shell keeps getting finished and a Korean cell maker now owns it outright, with a product line pointed at the fastest-growing corner of the American battery market. Fewer automotive jobs than promised, probably. But a mothballed 680-acre concrete slab would have been a worse outcome, and for a while that looked like the likeliest one.
Images Via: Wikipedia