GM's Sodium-Ion Battery Revolution: Powering AI Data Centers and the Grid

TL;DR

• GM is moving beyond EVs and into grid-scale energy storage, including a new sodium-ion battery program aimed at AI data centers and its own facilities.
• The company is partnering with Peak Energy and says the first GM-built cells should reach trial production in 2028 at its Battery Cell Development Center.
• GM is also pushing vehicle-to-grid software and other storage efforts, positioning itself as a broader energy company in the AI era.

GM’s latest battery push is less about cars and more about the power hungry infrastructure behind artificial intelligence. The automaker is developing a new sodium-ion chemistry for stationary storage, with early use cases spanning AI data centers, manufacturing plants, and grid support.

GM’s new energy bet

General Motors unveiled a broader energy storage strategy this week, and the centerpiece is a partnership with startup Peak Energy to co-develop sodium-ion battery cells for large-scale storage. GM says the chemistry is being tailored for stationary applications rather than vehicles, reflecting a growing view inside the company that the next big battery opportunity may be in the grid, not the driveway.

That strategy is backed by substantial investment. GM has already committed $900 million to commercialize new battery chemistries, a figure that includes a new battery development center. The company expects the first sodium-ion cells from this program to enter trial production in 2028 at that center.

Why sodium-ion matters for AI infrastructure

Sodium-ion batteries are attractive for stationary storage because they can offer lower cost and rely on more abundant materials than lithium-based systems, according to the coverage of GM’s announcement. Reported advantages also include safety and wide temperature tolerance, which are especially valuable for grid and backup-power applications.

The trade-off is energy density: sodium-ion systems typically take up more space than lithium-ion batteries, which makes them less suitable for compact consumer devices and some vehicle uses. But for data centers, substations, factories, and other high-power installations, space can be a manageable compromise if the chemistry delivers lower costs and reliable performance.

Data centers, factories, and the grid

GM is explicitly targeting AI data centers because their power demand is surging and utilities are under pressure to supply more stable energy. The company is also planning to use the batteries in its own manufacturing facilities first, which gives it an internal testbed for validating performance while reducing its own energy costs.

That approach fits a broader pattern in GM’s energy strategy: use internal operations as a proving ground, then scale outward to commercial and utility customers. GM is also working on repurposing used EV batteries for stationary storage with Redwood Materials, and it continues to develop lower-cost lithium iron phosphate, or LFP, cells with LG Energy Solution.

Vehicle-to-grid becomes part of the plan

Alongside the sodium-ion effort, GM announced expanded vehicle-to-grid capabilities that would let certain EVs send electricity back to the grid during periods of high demand. The company says this is part of a broader effort to help stabilize the power system while creating new value for EV owners.

GM is also building software and charging tools around that vision, including a unified charging experience in its app ecosystem. Together, those moves suggest GM wants to connect cars, homes, factories, and grid storage into a single energy network rather than treat them as separate businesses.

A bigger strategic shift for GM

The most notable part of the announcement may be what it signals about GM’s identity. Multiple reports describe the company as moving toward a role as a distributed energy player, not just an automaker. In that framing, EV batteries become one part of a much larger portfolio that includes stationary storage, bidirectional charging, and software-driven grid services.

For GM, the timing is deliberate. AI’s electricity needs are rising quickly, utilities are searching for flexible capacity, and battery makers are racing to find chemistries that can serve the grid more efficiently. GM is betting sodium-ion can carve out a meaningful niche in that market, even if the technology arrives first in pilot and trial deployments rather than mass commercial rollout.

What happens next

The immediate milestone is 2028, when GM expects trial production of its first sodium-ion cells. Until then, the company will be refining the chemistry with Peak Energy, expanding storage partnerships, and testing how its broader energy products fit into the needs of data centers, utilities, and industrial customers.

If GM succeeds, the company could emerge as one of the few automakers with a credible position in both mobility and grid infrastructure. That would make its battery strategy more than a side project—it would turn energy storage into a core business alongside vehicles.

AndroGuider Team

Articles written by the AndroGuider team. We try to make them thorough and informational while being easy to read.