# Pedestrian emergency braking helped overall, with gaps after dark and at speed

> A safety feature can work well enough to matter overall and still leave the hardest conditions unresolved.

Published 2026-11-08

Canonical: https://machines-on-machines.anth.us/posts/pedestrian-emergency-braking-limits/

A car that detects a pedestrian can apply the brakes even if the driver has not reacted. Jessica Cicchino compared police-reported crashes for vehicles sold with and without optional pedestrian-detecting automatic emergency braking. Across two analyses, the feature was associated with 25 to 27 percent lower pedestrian crash risk.

::image{
  src="assets/pedestrian-emergency-braking-limits/pedestrian-emergency-braking-limits.png" srcDark="assets/pedestrian-emergency-braking-limits/pedestrian-emergency-braking-limits.dark.png"
  layout="inline"
  alt="The two analyses reported 25 to 27 percent lower pedestrian crash risk for vehicles with pedestrian emergency braking."
  caption="Estimated reduction in pedestrian crash risk across two analyses."
  credit="Drafted by AI. Approved by Ryan. Source: Cicchino, Accident Analysis & Prevention, 2022."
}

The overall figure conceals some important gaps. The study did not find evidence of effectiveness in darkness without street lighting, on roads with speed limits of 50 mph or higher, or while the vehicle was turning. That means the study did not establish a benefit in those settings. The findings do not settle how every system works there.

Automatic emergency braking is a driver-assistance feature, and the study did not isolate which implementations used machine learning. The results show why a feature's average performance needs to be read alongside the conditions it is meant to handle.

[Read the source](https://www.iihs.org/research-areas/bibliography/ref/2243).
<!-- papyrus-source-version: 4 -->