There’s a federal law coming that could lock a wheelchair user out of their own vehicle every single morning. This disabled drivers impairment detection mandate is a looming issue that could trigger before they even get out of the driveway. Not because they’re drunk or impaired, but because a camera looked at their eyes and didn’t recognize what it was seeing.
That’s not a hypothetical. That’s where this is headed if nobody speaks up.
What Section 24220 Actually Is
In November 2021, Congress passed the Infrastructure Investment and Jobs Act. Most people heard about bridges and broadband. Buried in Section 24220 is something that got a lot less attention: a mandate requiring that all new passenger vehicles be equipped with “advanced drunk and impaired driving prevention technology” — passive systems that continuously monitor the driver and can prevent or limit vehicle operation if impairment is detected.
No breathalyzer. No officer. No due process. Just a camera watching your eyes, your head position, your blink rate — and software making a call.
The original deadline for NHTSA to finalize the rule was November 2024. That deadline passed without action. As of early 2026, the rule still isn’t finalized, but the implementation window is still targeting the 2027 model year. The mandate isn’t gone. It’s just delayed.
The Technology Isn’t Ready — NHTSA Said So Themselves
Here’s where it gets important. NHTSA’s own report to Congress acknowledged that no available technology currently meets the accuracy threshold required to deploy this system at scale. Not close. Not almost. Not ready.
And even if it were — even if the system hit 99.9% accuracy, which is better than any system currently in development — the math is damning. Americans drive hundreds of millions of vehicles across roughly three trillion miles per year. At 99.9% accuracy, NHTSA’s own analysis acknowledged that could mean millions to tens of millions of false flags annually. Millions of sober drivers prevented from starting their cars, or having their vehicles slow down without warning, because the software got it wrong.
That’s not a fringe concern. That’s the agency’s own conclusion. And it’s why the deadline slipped.
The public conversation has been mostly about privacy — who owns the data, whether the government can remotely disable your car, what insurers will do with the information. Those are real concerns. But they’re not the only ones worth having.
The Population That Gets Overlooked. Again.
Here’s the question I haven’t seen anyone ask: What happens to disabled drivers?
We’re talking about systems that primarily use driver-facing infrared cameras to monitor eye movement, pupil dilation, head position, and blink frequency. The logic is that an impaired driver will show irregular patterns across those metrics. That’s a reasonable premise — for the average driver.
But drivers with physical disabilities don’t present like the average driver. And a camera-based system has no way to know the difference.
Consider what these systems are designed to flag:
- Limited or restricted head movement
- Unusual or asymmetrical eye patterns
- Tremors or involuntary muscle movement
- Reduced blinking or abnormal blink rates
- Head position that doesn’t align with the camera’s expected forward-facing baseline
Now think about the conditions that produce exactly those characteristics in people who are completely sober, fully capable, and legally licensed to drive every day:
Parkinson’s disease. Multiple sclerosis. Cerebral palsy. ALS. Spinal cord injuries. Nystagmus. Bell’s palsy. Conditions affecting muscle tone, eye control, or head mobility. Veterans with traumatic brain injuries. People with prosthetic limbs whose driving posture differs from what the algorithm expects.
These aren’t edge cases. According to the CDC, roughly 61 million adults in the United States live with some form of disability. A meaningful portion of them drive, often with adaptive equipment that took enormous effort — and cost — to obtain. Their independence behind the wheel isn’t a convenience. In most parts of this country, it’s the difference between living a full life and being housebound.
A camera that interprets a tremor as impairment doesn’t know any of that. It sees a pattern that matches its training for “something is wrong” and acts accordingly. And unlike a breathalyzer — where you blow a 0.00 and you’re done — there’s no equivalent override built into these systems. There’s no mechanism being discussed, at least not publicly, for a driver to say: this is just how I move.
Nobody Is Advocating for This Community Right Now
The rulemaking process is where these concerns get addressed — or don’t. NHTSA is still in the process of defining exactly how this technology must perform, what thresholds it must meet, and what safeguards must exist. That process has involved automakers, safety organizations, privacy advocates, and consumer groups.
Disability advocates have not been a prominent voice in this specific conversation.
That’s a problem. Because by the time a final rule is published and automakers start building to it, the design decisions will already be locked in. Retrofitting accommodation for a population that wasn’t considered during development is always harder, more expensive, and less reliable than building it in from the start.
The questions that need to be on the table now:
- Will there be an exemption or calibration process for drivers with documented physical conditions?
- Who manages that process?
- Can a driver’s baseline be established at time of purchase and stored in the vehicle?
- What happens when someone is falsely locked out — what’s the override?
- Does the system require testing across diverse populations before deployment, including people with disabilities?
These aren’t unreasonable asks. They’re basic ones. And as far as the public record reflects, they’re not being asked loudly enough.
Why This Matters Beyond the Disability Community
The math problem with false positives isn’t just a disabled-driver problem. It’s a problem for tired parents, people recovering from eye surgery, anyone with bloodshot eyes on a long drive, anyone whose head naturally tilts. But for most of those people, a false flag is an inconvenience — frustrating, potentially dangerous in an emergency, but ultimately temporary.
For a driver with a permanent physical condition, the false flag isn’t occasional. It’s every single time they get in the car. The system will flag them on Monday morning the same way it flagged them on Friday afternoon. There’s no pattern it can learn that tells it this person is fine, this is just them. Not unless someone builds that capability in deliberately, with this population in mind.
That’s the difference between a bug that gets fixed and a feature that was never designed to work for you.
Drunk Driving Kills People. That Matters.
I want to be clear about something: the intent behind this mandate is legitimate. Drunk driving kills tens of thousands of people in this country every year. If technology can reduce that, it should. Nobody here is arguing for the right to drive drunk.
But a law designed to save lives shouldn’t silently create a class of drivers who are locked out of their vehicles every day for reasons that have nothing to do with impairment. The goal and the outcome have to be evaluated together. Good intentions and bad implementation aren’t mutually exclusive.
The rulemaking is still open. The standards are still being written. There’s still time for this to be done right. But only if someone makes the argument — loudly, specifically, and on the record — that disabled drivers exist, they drive, and they deserve to be designed for.
DC Auto Enhancement, LLC is an automotive customization and enhancement shop based in Greensburg, Indiana. If you found this worth reading, share it with someone in the rulemaking conversation.