Congress Ordered Cars to Stop Drunk Drivers in 2021. NHTSA Is Still Gathering Comments.

Five years. That's how long a dead family has been waiting for the federal government to do the thing Congress told it to do.

Section 24220 of the Infrastructure Investment and Jobs Act gave NHTSA a clean order: finalize a rule requiring impaired driving prevention technology in all new passenger vehicles. Three-year deadline. November 15, 2024. As of July 2026, NHTSA hasn't published a proposed rule, let alone a final one. Per their own February 2026 report to Congress, they're still analyzing comments from a January 2024 advance notice. Not a proposed rule. An advance notice asking whether they should maybe start thinking about proposing one.

That number falls out of the FARS data and basic multiplication. Across the 2014–2023 dataset, 20.0% of all drivers in fatal crashes tested positive for alcohol or drugs: 98,348 out of 490,736. Roughly 9,835 per year. Multiply by the 4.7 years since the IIJA became law. NHTSA has not performed this calculation in any of its three reports to Congress. Probably wise. It would look bad in a PowerPoint.

The law carries the names of Issam and Rima Abbas and their three children, killed by a wrong-way drunk driver in Kentucky. MADD and IIHS estimated the technology would save 10,100 lives annually at full fleet penetration. Even at half that optimism, roughly 23,700 people died preventable deaths while NHTSA sorted through public comments. NHTSA's explanation, repeated annually with the consistency of a liturgy: detection technology "around the legal limit continues to have an error rate that would be unacceptably high." False positives could strand sober drivers. So they keep studying.

Sports cars show the highest FARS impairment rate at 22.5%. Pickups sit at 20.1%. Vans clock the lowest at 18.1%, which tracks: family vehicles driven by people with somewhere to be sober. But the gap between classes barely matters. One in five drivers in a fatal crash is impaired, whether they're piloting a Corvette or a Caravan. That uniformity is exactly why Congress tried to mandate a fleet-wide fix instead of targeting individual models.

For contrast: NHTSA finalized FMVSS 141 in 2016, requiring EVs to make a fake noise below 18.6 mph so pedestrians can hear them. Affects less than 10% of registered vehicles. Start to finish, under four years. In June 2026, NHTSA requested OMB approval for yet another study, this one on a contextual Driver Monitoring System prototype. Still collecting data. A fake car noise got through the regulatory pipeline in under four years. Preventing 10,000 alcohol deaths annually is entering year five of asking people what they think.

What you can do: Check whether your vehicle has a driver monitoring system. Most 2024+ models with Level 2 ADAS include one. If yours doesn't, the technology Congress said should be mandatory by now isn't arriving anytime soon. NHTSA's comment docket, RIN 2127-AM50, remains technically open.

Limitations: FARS impairment data covers drivers in fatal crashes only, not all impaired driving incidents. "Any impairment" includes low-level drug positives that may not have caused the crash. Full fleet penetration of prevention tech would take a decade after rule finalization even in the best case. And the strongest counterargument deserves real weight: NHTSA's caution on false positives is a legitimate engineering concern. A system that kills a sober driver's engine on a highway at 70 mph is its own category of safety hazard. Mandating immature technology could create more body bags than it closes.