While much of the public debate about self-driving cars focuses on safety, a new national study from the University of California San Diego reveals Americans' doubts about driverless cars aren't just about the fear of a crash. Many Americans also fear the technology's economic ripple effects — especially job losses in driving and delivery work and the possibility that automated vehicles (AVs) could widen income inequality.
The study, accepted for publication in Transportation Research Part A: Policy and Practice, analyzed responses from 4,631 U.S. adults in the Pew Research Center's American Trends Panel, a nationally recruited survey. The study reveals a complex landscape of economic apprehensions that could impede widespread AV adoption if left unaddressed.
Among respondents, about 85% said widespread use of driverless cars would lead to job losses for ride-hailing, ridesharing and delivery drivers. More than 46% said driverless cars would increase the income gap between higher- and lower-income Americans (compared with about 6% who thought it would decrease income gaps). And more than 62% said they probably or definitely would not want to ride in a driverless vehicle.
"Driverless cars are often framed as an engineering challenge, but it's also a profound sociotechnical transition," said Behram Wali, lead author of the study and assistant professor in the Department of Urban Studies and Planning at the UC San Diego School of Social Sciences. "This study develops a new behavioral framework to reveal a critical tension: how Americans willingness to embrace driverless cars is directly tied to their fears of job loss and income inequality. These findings show many Americans are evaluating automated vehicles as a broader social and economic change — not just whether the technology works, but who benefits and who bears the costs."
How driverless cars may be AI's biggest trust test — and why many Americans say they're failing it
Driverless cars are one of the clearest real-world tests of AI because they use machine learning to interpret a messy, fast-moving physical environment — full of pedestrians, cyclists and unpredictable drivers — then translate those split-second decisions into the actions of a multi-ton vehicle.
The UC San Diego study suggests that trust is shaped not only by concerns about technical performance, but by concerns about system-level consequences. Concerns about job displacement and inequality echo broader debates about AI: it's not only whether AI can do a task, but who benefits, who bears the costs and how livelihoods may change.
A socio-technical divide among respondents
Notably, the research reveals a sharp divide in who is open to adopting driverless cars — and why. In the study, groups that were more aware of automated vehicles, used the internet more frequently, or had higher levels of education and income tended to be more willing to ride in driverless cars. At the same time, these same pro-tech groups were also more likely to believe the technology could worsen income inequality and disrupt jobs.
By contrast, lower-income respondents and people living in rural and non-metropolitan areas were less inclined to adopt driverless cars and expressed strong concern about negative economic impacts.
Building trust will take policy, not just technological advances
The study argues that a purely technical approach to automated vehicle adoption is unlikely to be enough to build public trust. Instead, Wali calls for a broader "socio-technical" approach — pairing technological development with proactive policy strategies that address economic anxieties and equity concerns.
"While conventional strategies such as increasing awareness and tech-savviness are helpful and necessary to boost AV acceptance, this study shows that such strategies alone cannot address the fundamental employment and economic concerns," Wali said. "We cannot afford a laissez-faire approach to AV regulation. Policymakers must ensure that underrepresented groups are not left behind."
Wali concludes that the question is no longer whether AVs will be deployed, but if and how can they be fairly and safely integrated into society.
"This study proposes critical policy interventions, including workforce protection through reskilling and upskilling initiatives, expanded social safety nets and programs to ensure equitable access across the geographic and social fabric of the nation," Wali said.
The paper draws on Pew Research Center's American Trends Panel, using data collected Nov. 1–7, 2021, which included questions about driverless vehicles and other advanced technologies, including AI.
