menu_open Columnists
We use cookies to provide some features and experiences in QOSHE

More information  .  Close

San Francisco is obsessed with the safest drivers — and ignoring the ones killing people

5 0
previous day

Waymo vehicles in a San Francisco lot await deployment in 2024. The debate over street safety in San Francisco seems to fixate on autonomous vehicles.

At around 9 p.m. on Feb. 27, a 2-year-old girl was walking in the crosswalk at Fourth and Channel streets in Mission Bay — one of San Francisco’s fastest-growing family neighborhoods — when a car turning right struck her and her mother. Bystanders screamed for the driver to stop. The child was pinned beneath the vehicle. She died at the hospital. Her mother was hospitalized.

Within days, two more pedestrian deaths followed: a man killed on the sidewalk in North Beach after an SUV reversed downhill at speed and a woman killed in a hit-and-run in the Outer Mission — at the same intersection where an 80-year-old had been killed just three months prior. 

Get Digital Access and Stay Informed With Trusted Local News.

Get Digital Access and Stay Informed With Trusted Local News.

The carnage hasn’t stopped.

Article continues below this ad

Last month, a Muni bus killed a pedestrian in San Francisco. And just this week, a driver killed a pedestrian in the South of Market area and fled the scene. Another fatality, another headline, another moment that we can’t afford to let pass without a structural response.

In the North Beach killing last month, police released the driver at the scene. A witness told Streetsblog that an officer said, “Don’t be too hard on yourself. It’s an accident. It happens.”

See more S.F. Chronicle on Google

“It happens” might be the closest thing San Francisco has to an official policy on traffic death.

This year, the driver who had plowed through a family of four at 75 mph in West Portal in 2024, killing them all, was let off the hook with a license suspension, two years of probation and 200 hours of community service.

Article continues below this ad

San Francisco adopted Vision Zero in 2014, promising zero fatalities by 2024. Instead, 43 people were killed in 2024 — the highest toll since 2005, exceeding that year’s homicides. Police traffic citations have collapsed by 95% over the past decade. Since Vision Zero began, 369 people have died on San Francisco’s roads. The city replaced the plan, then suspended its residential traffic-calming program due to budget cuts.

It’s against this backdrop that local leaders continue to fixate on autonomous vehicles.

Consider what happened when an autonomous vehicle killed a cat.

In October 2025, a Waymo vehicle ran over a beloved bodega cat in the Mission District. Within days, a supervisor held a rally and introduced a resolution calling for new robotaxi regulations. The story ran in dozens of national outlets, including two New York Times articles.

Even when I tried to explain the situation from a transportation engineering standpoint on “KQED Forum,” there was a focus on emotion rather than facts. I am a professor of transportation focused on roadway design and safety, and I discussed what happened in this situation — how the Waymo vehicle responded as it should have by minimizing risk.

Yet emotions have poured out for a cat that scurried under a car. Meme coins were created. In Los Angeles, people smashed Waymo vehicles while spray-painting “Justice for KitKat.”

I do not begrudge the cat lovers or mourners, nor do I dismiss legitimate questions about autonomous vehicles, but what I cannot dismiss is the asymmetry of the dialogue.

There were no hearings after a toddler died in a crosswalk. There were no emergency regulatory actions after repeated fatalities at known high-injury corridors. There have been no meaningful police enforcement responses despite a collapse in traffic citations. Yes, there are speed cameras now, but there’s also the San Francisco Police Department’s incongruous ticketing of “big, bad dangerous cyclists” instead of “overzealous drivers who keep killing pedestrians,” as someone wrote on Reddit.

This is not just inconsistent. It is backward.

California has one of the most intensive regulatory frameworks in the country for autonomous vehicles. The California Department of Motor Vehicles and Public Utilities Commission tightly controls operations. The U.S. National Highway Traffic Safety Administration requires detailed reporting of any crash involving injury or property damage within days.

For example, the federal government opened an investigation when a Waymo vehicle in Santa Monica struck a child — who sustained minor scrapes and went to school the same day. When a software issue caused Waymo vehicles to pass stopped school buses — causing zero injuries — the company recalled software from 3,000 vehicles.

Waymo recently published data from over 170 million fully autonomous miles, based on federally required reporting. The company’s vehicles show 92% fewer crashes involving serious or fatal injuries compared to human drivers on the same roads. 

In any other domain, for example, if this were the COVID vaccine, that would be treated as a breakthrough. Here, it is treated as a threat.

We are applying maximum scrutiny to a system demonstrating measurable safety gains, while applying minimal accountability to the status quo system that continues to produce predictable, preventable deaths. We are debating hypothetical risks while ignoring empirical outcomes; regulating performance at the margins while tolerating failure at the core.

If San Francisco were serious about safety, the focus would not be on constraining technologies that reduce harm. It would be on aggressively addressing the dominant source of that harm: human driving behavior, weak enforcement and an underbuilt street infrastructure that fails to support pedestrians and cyclists.

Guest opinions in Open Forum and Insight are produced by writers with expertise, personal experience or original insights on a subject of interest to our readers. Their views do not necessarily reflect the opinion of The Chronicle editorial board, which is committed to providing a diversity of ideas to our readership.

Read more about our transparency and ethics policies

Instead, we have inverted the problem. We have created a political and regulatory environment that preferences human drivers first, where the safest systems face the highest bar and the most dangerous ones are simply accepted as the cost of doing business.

Until that changes, the pattern will continue — and at the expense of the most vulnerable roadway users. Not because we lack the tools to prevent drivers from killing people — but because we are choosing to look in the wrong direction.

William Riggs is a professor at the University of San Francisco who works to help cities integrate emerging technologies into more sustainable, human-centered transportation systems.


© San Francisco Chronicle