Since global markets entered their second week of a tech-driven rout — with oil at $98 and AI infrastructure spending under the microscope — energy demand has become the defining pressure point of 2026. GM stepped into that pressure on Tuesday with one of the most ambitious pivots any legacy automaker has announced in the battery space.

What GM Actually Announced

Three things happened at GM's San Francisco event, according to reporting from The Verge, TechCrunch, and CNBC.

First, GM is activating vehicle-to-grid (V2G) capabilities for existing EV and home energy customers. Millions of EVs sitting in driveways carry enormous stored energy. GM wants to tap that — sending electrons back to the grid during peak demand rather than letting them sit idle.

Second, GM is partnering with Peak Energy, a Denver-based startup founded in 2023, to develop sodium-ion battery cells specifically designed for grid-scale deployments. The leadership team at Peak includes veterans of Tesla, Lockheed Martin, and Northvolt, according to CNBC.

Third, GM is rolling out a new feature to simplify public charging for EV owners — a secondary move that doesn't deserve top billing but signals the automaker is still fighting to make its EV ownership experience less painful.

Why Sodium-Ion Matters

The big technical story here is the battery chemistry itself. Sodium-ion batteries work similarly to lithium-ion, but swap out key materials. The result, according to TechCrunch's reporting on GM Vice President Kurt Kelty's comments: cheaper, longer-lasting cells that are significantly less prone to overheating.

The tradeoff is size and weight. Sodium-ion packs need to be physically larger to store the same amount of energy. For an EV, that's a dealbreaker. For a stationary grid storage system or a data center — where weight doesn't matter — it's a reasonable engineering compromise.

Peak Energy's sodium-ion systems require no active cooling and no fire suppression systems. As Paul Menson, GM's director of energy storage commercialization, told TechCrunch: "The hardest part to engineer is no part at all. Eliminate the part, eliminate the problem."

Lower upfront costs. Lower maintenance costs. No cooling infrastructure. For a data center operator staring down a $98-oil energy bill, that's an attractive pitch.

According to CNBC, GM's battery VP Kelty said sodium-ion has "the potential to reshape grid-scale energy storage" and can "operate across a wider range of temperatures and for more cycles" than lithium-ion.

What the Coverage Is Getting Wrong

Every outlet covering this story framed it as GM "joining" a race already underway. That framing undersells the ambition. Redwood Materials attached recycled EV packs to a Crusoe data center in Nevada. Ford repurposed some battery manufacturing capacity. Those are incremental moves.

GM is developing an entirely new battery chemistry from scratch for grid applications. According to TechCrunch, outside of China, no other automaker has announced plans to build sodium-ion cells. GM is attempting to run a different race entirely.

The coverage also glosses over the timeline problem. GM's sodium-ion cells won't enter trial production until 2028 at the company's Battery Cell Development Center. Customer deployment comes after that. This is a 2029-or-later story, minimum.

The AI energy crisis is happening NOW. Oil is at $98 NOW. The V2G announcement gets you something in the near term — but the sodium-ion bet is a multi-year gamble.

The Fair Concern: Is This Just Capital Allocation Theater?

Skeptics have a legitimate question here. GM has committed $900 million to commercialize new battery chemistries, per TechCrunch — but the company declined to say how much of that is specifically allocated to this sodium-ion effort. That's a red flag for anyone watching how corporations chase hot narratives.

GM's EV sales have cooled. The automaker has been trying to break into the energy storage market for nearly four years, according to The Verge, without a landmark commercial success to show for it. There's a credible argument that this announcement is timed to capture AI-energy hype at a moment when investors are rewarding companies that position themselves near the data center buildout story.

That concern is fair. But it doesn't make the underlying technology wrong. Sodium-ion's cost and safety profile genuinely suits stationary storage. Peak Energy exists and has already demonstrated working systems. The Battery Cell Development Center is a real facility. The technology is real — the timeline and execution risk are the legitimate question marks.

What This Means for Regular People

If you own a GM EV, the V2G activation is the only thing happening in your timeframe. You could theoretically earn credits or reduce your energy bill by letting the grid draw from your battery during peak demand.

For everyone else — taxpayers, ratepayers, consumers — the broader picture is this: the AI buildout is consuming electricity at a scale that is outpacing grid expansion. That is pushing energy costs higher across the board. Oil at $98 makes everything more expensive.

GM's sodium-ion play, if it works, could provide cheaper, safer grid storage by the early 2030s. But utilities need solutions in the next 18 months, not the next 18 quarters.

The energy crisis created by AI infrastructure spending is real and urgent. GM's answer is ambitious and scientifically sound. The gap between those two facts is where regular people are going to feel the pain.