The Promise of Self-Driving Cars: A Road to Safety

It’s a hard truth to swallow, but most of us are probably not as good drivers as we think. Human driving is inherently flawed. We are prone to distractions, fatigue, anger, and, tragically, driving under the influence. Even when fully alert, our brains, evolved for a simpler world, struggle with the demands of high-speed vehicular operation. The delay between perception and reaction while driving at 60 mph can be as long as 2.5 seconds, meaning a car travels the length of two basketball courts before the driver can even begin to brake.

This inherent fallibility results in a devastating toll on human life. Globally, nearly 1.2 million people die in road crashes each year, a number equivalent to nine fully loaded jumbo jets crashing every day. In the United States alone, the government estimates nearly 40,000 traffic fatalities in 2024. This translates to a busload of people perishing every 12 hours.

However, a potential solution is emerging, one that offers the promise of significantly safer roads: self-driving cars. These autonomous vehicles have capabilities that human drivers simply cannot match. They don’t require sleep, are immune to anger, and never drive under the influence. Their advanced computing power allows them to make high-speed decisions with greater precision and speed than any human brain.

Consider the time an average American spends driving: roughly three years of their life. If autonomous vehicles could take over, imagine the possibilities – the time reclaimed for leisure, work, or simply relaxation.

But the most significant benefit lies in the potential to save lives. Recent data from Waymo, an autonomous vehicle company, suggests that these savings could be substantial.

Waymo conducted a peer-reviewed study, soon to be published in the journal Traffic Injury Prevention, analyzing the safety performance of its autonomous vehicles over 56.7 million miles driven in Austin, Los Angeles, Phoenix, and San Francisco. Importantly, these vehicles operated without a human safety driver present to intervene in emergencies. The study compared this data to human driving safety over a comparable distance and on similar types of roads.

The results, representing arguably the most comprehensive research on self-driving car safety to date, were remarkable. Waymo’s self-driving cars demonstrated a significantly lower crash rate compared to human drivers. Specifically, researchers projected that if human-driven cars had covered the same 56.7 million miles, there would have been an estimated 181 additional injury crashes, 78 additional airbag deployments, and 11 extra serious-injury crashes.

Extrapolating this data across the 3.3 trillion vehicle miles driven by humans annually in the US paints an even more compelling picture. If the 85 percent reduction in serious crashes observed in the Waymo study extended to fatal crashes – a significant assumption given the limited number of fatal events in the study – approximately 34,000 lives could be saved each year. To put this in perspective, that’s five times the number of Americans who died in the Iraq and Afghanistan wars combined.

Of course, it’s essential to acknowledge the caveats surrounding the Waymo study and the challenges that lie ahead. Due to the relatively low frequency of serious injury crashes, even 56.7 million miles may not be sufficient to definitively prove that robot drivers significantly reduce their occurrence. Further data collection is crucial.

Furthermore, Waymo’s testing primarily occurred in warm, sunny locations within geofenced areas meticulously mapped by the company. The performance of these vehicles in challenging environments, such as the snowy streets of Boston during winter, remains uncertain. It’s also important to remember that this study was conducted by Waymo, although it underwent peer review by external experts.

Even if we fully embraced AI drivers, producing enough autonomous vehicles to replace the existing fleet of human-driven cars and trucks would be an immense undertaking.

Despite these challenges, the potential benefits are undeniable. The data strongly suggests that autonomous vehicles offer a pathway to dramatically reducing road fatalities. Given the staggering human cost of car accidents, slowing down the development and deployment of autonomous vehicles could be seen as costing lives.

Unfortunately, public perception of self-driving cars has been negatively influenced by unusual, high-profile incidents, while the everyday carnage caused by human drivers is often overlooked. This discrepancy stems from psychological biases such as availability bias, which leads us to overestimate the risk of events that are easily recalled, and base-rate neglect, where we ignore the underlying frequency of events. This skewed perception is often reflected in news coverage, contributing to a negative public sentiment.

This growing skepticism has, in some cases, even led to vandalism targeting autonomous vehicles. Moreover, the widespread adoption of self-driving technology would undoubtedly disrupt the livelihoods of the nearly 5 million Americans employed in driving-related occupations, presenting significant economic challenges.

However, the potential to save 34,000 lives annually is a compelling argument for pushing forward. After a century of relying on human drivers, perhaps it’s time to consider a new approach behind the wheel.