← The Crash Report
Investigation

NYC Put Speed Governors on 700 Cars; Speeding Dropped 64%; Nobody Sells Them

New York City installed a device on 700 fleet vehicles that limits the accelerator when the car exceeds the posted limit by more than 11 mph. A US Department of Transportation evaluation found speeding dropped 64 percent.[1] Even among the worst habitual offenders, the reduction was 49 percent. Each unit costs a few hundred dollars per vehicle, does not touch the brakes, and simply makes the gas pedal push back once GPS confirms the car is over the limit.

7,608
estimated annual US deaths preventable if ISA were standard on all vehicles

The European Union has required intelligent speed assistance on every new car sold since July 2024,[2] but in America, zero automakers install it. NHTSA attributes 29 percent of all traffic fatalities to speeding, 11,887 dead in 2023.[4] Run NYC's 64 percent through that number and the product is 7,608 funerals that did not need to happen, roughly 21 per day.

ISA works because it targets the behavioral layer, not the crash layer. Seatbelts and airbags try to keep you alive after you have already decided to die at 85 miles per hour. AEB intervenes in the last 1.5 seconds before impact, a window where the outcome was already decided by the speed the driver chose. ISA prevents the condition that makes the crash unsurvivable in the first place, upstream of every other safety system in the vehicle.

The Ford Mustang ships with 486 horsepower and a 6.02 fatality rate per 100 million vehicle miles traveled, six times the national average and the second-highest death rate in the FARS database among vehicles with more than 200 recorded fatalities.[3] Of the Mustang's 2,739 deaths over the past decade, 78 percent involved sober drivers. When impairment is not the explanation and vehicle design is not markedly worse than peers, speed is the only variable left.

The counterargument is obvious: drivers will override it. NYC tested that hypothesis on municipal employees with documented histories of habitual speeding. Repeat offenders still cut speeding by 49 percent. Seatbelts followed the same trajectory, taking decades of mandates before compliance hit 90 percent but saving lives from the first day the law took effect.

New York State just included the "Stop Super Speeders" law in its 2026 budget, requiring ISA for any vehicle caught on speed cameras 16 or more times in 12 months.[1] At the federal level, NHTSA has no ISA rulemaking and no timeline. The arithmetic is not ambiguous: if speeding kills 12,000 per year and a device costing less than a set of floor mats cuts that rate by 64 percent, then 7,600 annual deaths are a policy choice.

What you should do

If you drive a vehicle with more than 300 horsepower, your fatality risk per mile is statistically higher than the fleet average. You cannot buy ISA from a US dealer, but aftermarket GPS-based speed alert systems shift the behavioral default. Check NHTSA's speed-campaign page for enforcement calendars in your state.

Sources & References

  1. Streetsblog NYC, “City’s In-House Program Proves Speed Governors Work,” May 12, 2026. Citing US DOT evaluation of NYC DCAS ISA pilot program (Oct 2024). streetsblog.org
  2. European Commission, Regulation (EU) 2019/2144 on type-approval of motor vehicles, mandating ISA on all new vehicles sold from July 6, 2024. eur-lex.europa.eu
  3. NHTSA, Fatality Analysis Reporting System (FARS), 2014–2023. Per-model death rates calculated from FARS fatal crash data, NCES fleet estimates, and NHTS annual VMT. nhtsa.gov
  4. NHTSA & NSC, “Speeding,” Injury Facts. Speed-related fatalities: 11,887 in 2023 (29% of all traffic deaths). injuryfacts.nsc.org
  5. NHTSA, “Almost One-Third of Traffic Fatalities Are Speed-Related Crashes,” speed campaign launch (2022). nhtsa.gov

Source: NHTSA FARS 2014–2023; NYC DCAS ISA evaluation (US DOT, 2024); NHTSA speed-fatality statistics 2019–2023. Fatality rate estimates use VMT derived from NHTS survey data and carry ±15% uncertainty for low-volume models. ISA effectiveness extrapolation assumes NYC fleet results are transferable to the general driving population; actual national effectiveness may vary. See methodology for caveats.