Only one person died of a lithium-ion battery fire in New York City last year. She was 76 years old, she was in the bathroom of a Queens pizzeria, and it was the Fourth of July. The Fire Department of the City of New York later traced the ignition to an e-mobility device charging just outside the bathroom door.
That number — one death — is the headline. In 2023, the number was 18. Through a combination of inspections, summonses, a $1 million public-awareness campaign, and a city law that took effect in September 2023 banning the sale of any e-mobility device without UL safety certification, New York City cut its lithium-ion battery fire deaths by 94 percent in two years. The new research published in The Guardian on May 11, showing that British fire brigades are now responding to a lithium-ion battery fire every five hours, describes a curve America has already been climbing. The difference is what the rest of America is going to do about it.
What the FDNY data actually says
Since 2022, lithium-ion batteries have caused more than 800 fires, 30 deaths, and 400 injuries in New York City alone, according to FDNY records and reporting compiled by the NFPA Journal. The trend was steeply upward for several years before it reversed. The city saw 268 lithium-ion fires in 2023 and 277 in 2024, with fatalities falling from 18 to six over the same period. In the first quarter of 2025, structural fires caused by lithium-ion batteries rose 53 percent compared with the same period the previous year. The death count went the other way: zero through Q1, one for all of 2025.
Most of those fires were not Teslas. Most of them were not full electric vehicles at all. The biggest single category, in New York and everywhere else fire departments break the data down, is micromobility: e-bikes, e-scooters, hoverboards. The most dangerous configurations are uncertified imports, aftermarket conversion kits, and second-hand batteries paired with chargers they were never designed to work with. The U.S. Consumer Product Safety Commission has noted that “most” or “nearly all” micromobility platforms being sold in the United States are imported from China — the same finding the UK National Fire Chiefs Council has cited in its own warnings.
The bill the House passed and the rule the CPSC then withdrew
In April 2025, the U.S. House of Representatives passed the Setting Consumer Standards for Lithium-Ion Batteries Act, H.R. 973, on a 365-42 vote. The bill, introduced by Representative Ritchie Torres of New York and co-sponsored by Andrew Garbarino, Yvette Clarke, and Nick Langworthy, would require the Consumer Product Safety Commission to adopt three voluntary industry safety standards as binding federal rules within 180 days of enactment: UL 2271 for the lithium-ion batteries in e-mobility devices, UL 2272 for the platforms themselves, and UL 2849 for the full electrical system of e-bikes. DoorDash supported the bill. So did the International Association of Fire Chiefs. The Senate version, S. 389, has bipartisan sponsorship from Senators Kirsten Gillibrand, Chuck Schumer, Marsha Blackburn, and Deb Fischer.
On May 8, 2025, President Trump fired three of the five sitting CPSC commissioners. On May 13, the reconstituted CPSC took an emergency vote to withdraw a proposed federal rule that would have established a safety standard for lithium-ion batteries used in micromobility products. The legislative push remained alive in the Senate. The federal regulatory floor that the House bill was meant to build on no longer existed.
In the absence of federal rules, individual states and cities are setting their own. New York City’s 2023 law was the first. San Francisco, Washington, D.C., and Denver have followed. California’s 2026 e-bike legislation includes battery safety provisions that are expected to influence other state codes. Outside those jurisdictions, an American consumer still has substantial legal protection from outright counterfeit batteries because Amazon and Walmart have tightened seller requirements, but the federal hard floor of “these batteries must meet UL 2849 to be sold in the United States” simply does not exist.
The training picture is worse than the regulation picture
A 2023 national study cited by New York State Senator Tom O’Mara found that 40 percent of U.S. first responders had no training in handling electric vehicle battery fires. New York State passed bipartisan legislation in 2024 — S5848 in the Senate, A247 in the Assembly — to create a state-funded EV emergency training program for first responders. Governor Kathy Hochul vetoed it, citing the absence of funding in the current financial plan.
The National Fire Protection Association offers a free training program for firefighters, police, and EMS personnel through the federal Alternative Fuels Data Center. Most of the practical work, however, is being done by independent trainers. The most prominent of them is Patrick Durham, a former EV battery engineer who is now a volunteer fire captain in Troy, Michigan, and runs a private training operation called StacheD Training that has become one of the most respected on the circuit. Jim Stevenson, a rural Michigan fire chief who has taken Durham’s course, summed up the practical reality this way: at most full EV fires, the only honest response is to set up a perimeter, protect adjacent structures, and let the battery pack burn out under controlled conditions. To the public watching, he noted, that looks like firefighters standing on the side of the road doing nothing. There is often nothing better to do.
Why these fires don’t behave like other fires
A conventional vehicle fire is a contained chemical reaction. Pull the line, knock down the flames, ventilate the smoke, write the report. A lithium-ion battery fire is something else. Inside the battery pack, individual cells generate their own oxygen as they fail. The reaction is self-sustaining. Cells reach 2,000 degrees Fahrenheit and propagate to adjacent cells faster than firefighters can cool them. A 90-kilowatt-hour EV battery, as a recent Scientific American piece noted, carries roughly the chemical energy of 170 pounds of TNT.
The reignition risk is now part of every responder protocol. In one widely cited 2020 Chevrolet Bolt fire, firefighters spent an hour extinguishing what looked like a closed-out fire. It reignited an hour later. It reignited a third time after the vehicle had been towed to a Chevrolet dealership service bay. GM’s recall of every Bolt EV ever produced over LG Energy Solution battery defects eventually cost the company about $1.8 billion. Hyundai’s parallel Kona Electric recall over the same LG cells cost roughly $900 million.
The International Association of Fire Chiefs now recommends 24-hour thermal monitoring of any vehicle that has experienced battery damage, isolation distances of at least 50 feet from buildings and other vehicles, and explicit briefings to tow operators about the ongoing reignition risk. The U.S. Fire Administration’s protocol explicitly warns that fire blankets, often marketed as a residential solution, have been tested by the Fire Protection Research Foundation and UL’s Fire Safety Research Institute and found to delay rather than stop thermal runaway. Once the blanket is removed, the oxygen returns, and the fire can flare back.
Water is still the primary suppression tool, but a lot more of it is required. A typical EV battery fire consumes 3,000 to 8,000 gallons of water, depending on pack size and access, compared with roughly 500 gallons for a conventional gasoline vehicle.
None of this means EVs are more dangerous than the gasoline cars they replace. The reverse is true. Industry data suggests that an internal combustion engine vehicle is roughly 60 times more likely to catch fire than an EV on a per-vehicle basis. The asymmetry is not in how often EVs ignite. It is in what happens after they do.
What this looks like in a real American garage
Picture a normal American household in 2026. A Ford F-150 Lightning or a Tesla Model Y is plugged into a Level 2 charger overnight. An older Lectric or Aventon e-bike that the parents bought during the pandemic hangs on a wall hook nearby, also charging. The kids have two Razor or Hover-1 e-scooters plugged into a shared power strip in the laundry room. On the kitchen counter, a vape charges next to two phones. In the basement, an older laptop whose battery has started to swell sits in a drawer because nobody has gotten around to recycling it.
The Tesla and the Lightning are the most expensive items on that list. They are also the least statistically likely to be the source of a fire. The risk concentration sits in the cheap, the converted, the aftermarket, and the forgotten. The systemic problem is not any single device. It is that the volume of lithium-ion cells inside the average American home has grown by roughly an order of magnitude in five years, and the federal regulatory framework, the firefighter training infrastructure, and the insurance industry have not kept up.
The insurance gap nobody talks about
If a lithium-ion battery ignites in an American garage, the insurance reality is messier than most homeowners realize.
When the source of the fire is a passenger EV, two policies are typically in play. The damage to the car itself falls under your auto comprehensive coverage. Damage to the garage structure, the house, and any personal property inside falls under your homeowners policy. In 2026, the average replacement cost of a high-voltage EV battery pack runs between $5,000 and $18,000 depending on the model, which is why insurers often total an EV even when the fire was contained.
When the source is an e-bike or an e-scooter, the picture gets uglier. Standard HO-3 homeowners and HO-4 renters policies in the United States contain a “motorized vehicle” exclusion clause. Coverage practice varies by carrier and by state: some policies treat low-power e-bikes the same as conventional bicycles, while others exclude any motorized two-wheeler outright. Even when the device itself is technically covered, claims can be denied on negligence grounds if the insurer can show that the owner used a non-OEM charger or a non-certified aftermarket battery — the exact configuration the FDNY and the UK fire chiefs both identify as the highest-risk category. The motorized-vehicle exclusion can also be triggered if the e-bike was used for delivery work, which is treated as commercial use even if the rider only ran the occasional DoorDash or Uber Eats shift on the side.
The honest summary is that homeowners insurance probably covers a lithium-ion fire that starts in your house. It may not cover the device that started it, and it may push back hard if the device was a converted Amazon-marketplace e-bike with a battery that did not ship from the original manufacturer. The policy language at issue was written for older categories of risk. It is rarely specific to lithium-ion.
What you can actually do
Buy e-bikes and e-scooters certified to UL 2849 for the full electrical system and UL 2271 for the battery alone. As of 2026, the major U.S. brands have largely converged on the standard. Lectric eBikes, the largest direct-to-consumer e-bike seller in North America, is certified across its full lineup. Aventon’s newer Sinch.2, Level.2, and Pace models carry the certification. Velotric’s entire lineup is certified. Trek, Specialized, Giant, and Cannondale have certified across their e-bike offerings. Rad Power Bikes certified its bikes to UL 2849 starting in September 2023 but filed for Chapter 11 bankruptcy protection in December 2025 and is currently restructuring. The greatest residual consumer risk is unbranded direct-from-factory imports sold through marketplace platforms, which Amazon and Walmart have tightened against but not eliminated.
Use only the charger that shipped with the device. Charge in a garage or on a balcony if you can, not in a hallway that blocks your exit if something goes wrong. Treat any visibly swollen battery as hazardous waste and dispose of it through your county’s household hazardous waste program, not in the regular trash. Call your insurance agent and ask in writing whether your e-bike, your home battery storage, and your kids’ e-scooters are covered under your existing policy. If they are not, a scheduled personal property endorsement is usually available for a modest annual premium.
And if a lithium-ion device does catch fire in your home, evacuate first and call 911 immediately. Tell the dispatcher it is a lithium-ion battery so the dispatcher can roll the right equipment and brief the responding company. Do not move the device. Do not try to smother it with a blanket. Do not use a kitchen extinguisher and hope. Get out. The fire is not the same kind of fire your parents taught you about, and it will not respond the same way.
What the curve actually says
The headline from London is not that EVs are dangerous. The American data, on a per-vehicle basis, says the opposite. The headline is that the volume of lithium-ion cells in American homes has grown faster than the federal regulatory framework, the firefighter training infrastructure, and the insurance industry can absorb. New York City built a response. It worked: lithium-ion battery deaths in the country’s largest city went from 18 in 2023 to one in 2025. Most of the rest of the country has not built that response, and the federal floor that the House voted to build in April 2025 was knocked down at the CPSC five weeks later.
The honest version of the warning is not that your Tesla is going to burn down. It is that the cheap e-bike on the wall hook, the older converted scooter in the hallway, the swollen laptop in the drawer, and the off-brand replacement battery you bought through an Amazon storefront last year are the actual fire risks; that the average American homeowners policy was not written with any of them in mind; and that whether your fire department knows how to handle the result depends a lot on the ZIP code where you live.
None of that is reason for panic. It is reason to know what is charging in your house tonight, and to make sure your insurance agent and your local fire chief know it too. New York City showed that the curve bends when the system catches up. The question is which other American cities are willing to do the work.
