Navigating Tomorrow's Streets: Unpacking What Truly Drives Robotaxi Adoption in San Francisco's Urban Jungle
The streets of San Francisco are increasingly becoming a testbed for the future of urban mobility, where traditional taxis and private cars share the road with a new breed of vehicle: the robotaxi. These shared autonomous vehicles (SAVs), which operate without the need for a human driver, represent a significant technological leap, promising to revolutionize how people move through cities. Companies like Waymo, owned by Alphabet, have already deployed these driverless services in San Francisco and other U.S. cities, transforming what was once a futuristic concept into a commercial reality. But for these innovations to truly integrate into daily life, understanding what motivates people to accept and use them is crucial. A recent study, "Riding Into the Future: What Drives the Use of Robotaxis in San Francisco?", delves into this very question, exploring the factors that influence individuals' intentions to use, and actual use of, robotaxis among residents of San Francisco and San Jose Metropolitan Statistical Areas (MSAs).
The Evolution of Mobility: From Concept to Reality
Robotaxis are not just another car; they are SAE Level 4 vehicles, meaning they are highly automated and require no human interaction in most circumstances, capable of operating entirely in self-driving mode. Their emergence signifies a shift not only in technology but also in mobility behavior, moving from car ownership towards shared services. While academic interest in autonomous vehicles (AVs) has soared, most previous research relied on hypothetical scenarios, showing pictures or detailed descriptions to respondents who had little or no real-world experience with these technologies. This created a "practical-knowledge gap" and a "theoretical research gap," as little was known about the factors driving acceptance and adoption where these services were commercially available and users had actual exposure. This study specifically addresses this gap by focusing on San Francisco residents who have at least heard of, seen, or even taken a ride in a robotaxi.
Furthermore, robotaxis are uniquely positioned at the intersection of three major technological domains: automated driving, robotics, and artificial intelligence (AI). Recognizing this convergence, the study adopted an interdisciplinary approach, drawing from literature streams in all three areas to develop its hypotheses.
The Unified Theory of Acceptance and Use of Technology (UTAUT2): The Study's Compass
To explore user behavior, the researchers applied an extended Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) framework. The UTAUT2 model is a well-established tool in social and behavioral sciences, known for its high predictive power in explaining why consumers adopt new technologies. This study expanded the model by including two additional factors: compatibility and personal innovativeness.
Compatibility refers to how well an innovation, like a robotaxi, aligns with an individual's existing values, needs, and past experiences. For instance, if a robotaxi fits seamlessly into someone's daily travel routines, they are more likely to use it.
Personal Innovativeness (PI) describes an individual's willingness to try out any new technology. This characteristic often plays a strong role in the early adoption of radical technological innovations.
The study also aimed to address the "intention-behaviour gap," examining not just if people intend to use robotaxis, but if that intention actually translates into actual use behavior, measured by how frequently they use the service. This is a significant advancement over previous studies that primarily focused on behavioral intention alone.
Peering into San Francisco's Mobility Landscape: How the Study Unfolded
The research team gathered data from 517 valid San Francisco and San Jose MSA residents through an online survey platform, Qualtrics, between January and March 2025. To ensure the participants had a minimum level of familiarity with robotaxis, only those who had at least heard of "autonomous taxis" were included. The survey deliberately used the term "autonomous taxi" to remain neutral and avoid biases associated with specific brand names like Waymo.
The survey was divided into three sections: usage and travel behavior, the extended UTAUT2 items (rated on a seven-point Likert scale), and socio-demographic information. A rigorous data cleaning process was implemented to remove careless respondents, ensuring the quality and reliability of the data.
The data analysis involved two main steps: confirmatory factor analysis (CFA) to validate the survey items and the theoretical constructs, and partial least squares structural equation modeling (PLS-SEM) to test the hypothesized relationships between the various factors and the intention to use robotaxis. A multi-group analysis (MGA) was also performed to see how different socio-demographic factors and mobility behaviors influenced these relationships.
Unveiling the Drivers: Key Findings on Robotaxi Acceptance
The study's findings paint a nuanced picture of robotaxi acceptance in San Francisco:
Experience Levels: Among the respondents, almost 15% had already used a robotaxi. More than half had seen them on public roads but hadn't taken a ride, and about a third had heard of them but never seen them. This demonstrates a growing, though still nascent, level of public exposure.
Prevailing Preferences: Despite the presence of robotaxis, traditional transport preferences still largely prevail in San Francisco. Private car use remains the most dominant mode, with two-thirds of respondents using their private car at least three times a week. Public transport use was relatively low, and car-sharing was the least used non-autonomous mode. However, ride-hailing services are more widely adopted, with 35% using them 1-3 times per month and 14% weekly.
Strongest Predictors of Intention: The study found that seven out of nine hypothesized factors significantly influenced behavioral intention to use robotaxis. The strongest positive significant path coefficients were observed for:
Personal Innovativeness (PI): Individuals who are eager to try new technologies are most likely to intend to use robotaxis. This suggests that "innovative and tech-savvy individuals" are the primary early adopters.
Social Influence (SI): If "people who are important to me think that I should use an autonomous taxi," this significantly increases an individual's intention to use it.
Hedonic Motivation (HM): The perception that traveling in a robotaxi would be fun, enjoyable, and comfortable strongly influences intention to use.
Other factors with significant positive effects included Performance Expectancy (PE) (perceiving robotaxis as useful and safe), Facilitating Conditions (FC) (having the knowledge and resources to use them), Habit (HH) (robotaxis becoming a routine), and Compatibility (COM) (fitting into existing travel routines).
Factors with No Significant Effect: Interestingly, Effort Expectancy (EE) (ease of use) and Price Value (PV) (perceived reasonable pricing) did not show a significant effect on the intention to use robotaxis in this study. This finding for effort expectancy aligns with some prior research that suggests it may not always be a strong predictor in the context of autonomous technology. For price value, previous studies have shown mixed results, ranging from positive to no effect or even negative effects.
Uncertainties and Disagreements:
Price Value: A notable finding was that almost half (43%) of respondents were undecided about whether robotaxis offered "good value for money". This "high level of neutral responses" suggests public uncertainty about the economic benefits.
Social Support: The item "People who are important to me think that I should use an autonomous taxi" received the lowest level of agreement and the second-highest level of neutral responses (36%).
Habit Formation: Questions related to the repeated use of robotaxis, becoming a habit, or integrating them into daily life elicited strong disagreement. This implies that for many, robotaxis are not yet seen as fulfilling a necessary daily travel need, possibly viewed as a novelty or "tourist gimmick" rather than an inclusive mobility option.
Authority Support: About a third of respondents were uncertain about local authority support for robotaxis, reflecting ongoing regulatory uncertainties. However, the low level of disagreement suggests authorities are not perceived as hindering progress significantly.
From Intention to Action: Actual Use Behavior: Critically, the study empirically supported that a higher behavioral intention to use robotaxis has a positive and significant effect on actual use behavior (i.e., the frequency of robotaxi use) for those who have already taken commercial rides. This is a major contribution, helping to close the theoretical intention-behavior gap in autonomous mobility research.
The Nuances of Acceptance: Moderating Effects
The multi-group analysis revealed how socio-demographic factors and mobility behaviors can influence the strength of these relationships:
Age: For younger respondents (under 45), both effort expectancy (ease of use) and habit were more important in driving their intention to use robotaxis.
Gender: Male respondents showed a significantly higher effect of price value on their behavioral intention compared to female respondents.
Prior Mobility Habits: Car-sharing and ride-hailing users (who are generally more open to new services) were more concerned with effort expectancy (ease of use). For individuals who frequently use private vehicles, social influence was a key factor in increasing their intention to use robotaxis; peer recommendations played a significant role for this subgroup.
Paving the Way: Practical Implications for Tomorrow's Mobility
These findings offer valuable insights for robotaxi operators, policymakers, and manufacturers aiming to increase adoption:
Target the Innovators: Operators should continue to focus efforts on innovative and tech-savvy individuals, as they are the most likely early adopters and can help foster initial growth.
Highlight Unique Benefits: Communication strategies should emphasize the joy, comfort, privacy, and efficiency of robotaxis, highlighting aspects like not needing to talk to a driver, enjoying personal music, or working during a trip. This taps into the strong influence of hedonic motivation.
Strategic Conversion: There's significant potential to convert regular ride-hailing customers to robotaxi users, given their openness to shared mobility services.
Community Engagement: To shift robotaxis from being perceived as "expensive gadgets" or "tourist gimmicks," operators and policymakers should engage with communities and specifically target individuals with higher mobility needs, such as the physically handicapped, to showcase robotaxis as a truly useful and valuable option. This can improve the overall perception of AVs.
Address Value Perception: The high level of uncertainty regarding "value for money" indicates an opportunity for targeted behavioral and policy interventions to persuade residents of the economic benefits of robotaxis.
The Road Ahead: Acknowledging Limitations and Future Directions
While groundbreaking, this study has limitations. Its findings are primarily generalizable to the San Francisco and San Jose MSAs, not necessarily to wider populations or other cities due to differences in travel patterns, urban forms, and cultural contexts. The use of a non-probability quota sample means there could be inherent biases, and priority populations might be underrepresented. Future research could expand the geographical scope and employ probability sampling for broader representativeness.
Furthermore, relying solely on survey data presents a limitation; future studies should integrate survey responses with naturalistic robotaxi driving data to gain a more comprehensive understanding of actual usage patterns. The study also suggests exploring alternative theoretical models, such as the AI Device Use Acceptance Model (AIDUA), which may be more specifically tailored to AI-enabled mobility, and to consider the influence of culture and tradition, which can significantly impact robotaxi acceptance.
In conclusion, this study offers pioneering empirical evidence from a region where robotaxis are a commercial reality, illuminating the complex interplay of factors that drive their acceptance and use. By identifying personal innovativeness, social influence, and hedonic motivation as the strongest predictors, and by confirming the critical link between behavioral intention and actual usage, it provides a vital compass for steering the future of autonomous mobility. As robotaxis continue to integrate into our urban fabric, understanding these human dimensions will be paramount to unlocking their full potential and ensuring a future where cutting-edge technology truly serves the needs of all city dwellers.
Autonomous Vehicle Researchers:
Dr. Sina Nordhoff: Post-doctoral researcher at the Institute of Transportation Studies UC Davis.
Dr. Jie Zhang is one of the authors of a study conducted by King's College London which found that autonomous driving pedestrian detection systems were less accurate at detecting children and people with darker skin tones.
Dr. Carlotta Berry is a robotics expert and engineering professor at the Rose-Hulman Institute of Technology. She founded NoireSTEMinist in 2020, an educational consulting firm promoting diversity in STEM. Berry is actively involved in robotics programs and authors romance novels featuring Black STEM characters, aiming to make STEM accessible to all.
