Analysis · Technology
Supporting Driverless Cars Is Different from Wanting a Ride
By AWEI · AI-compiled · Published · Analysis prepared · 1 source · doi.org
An official can support the development of driverless cars while remaining reluctant to ride in one. In a 2021 study by Michael C. Horowitz and Lauren Kahn, 39% of respondents scored higher on policy support than on personal willingness to use autonomous vehicles. That gap matters when interpreting enthusiasm for AI adoption.
The researchers used a CivicPulse survey conducted in October 2020 with 690 U.S. local officials. Of these, 555 completed the full survey. Questions covered autonomous vehicles, autonomous surgery and other AI applications. The analysis used probability weights to better represent U.S. local officials, alongside regression models accounting for individual and community characteristics.
The revealing methodological choice was to ask separately about supporting autonomous vehicle development and personally taking a ride. On the researchers’ comparison measure, 51% showed no gap, while 10% expressed greater willingness to use the technology than to support its development. A single approval question would have missed these distinctions.
What explains the gap remains unresolved. The authors suggest that officials might be open to permitting development while doubting whether vehicles are ready for personal use. Alternatively, broad policy questions might overstate their enthusiasm. These are interpretations, not demonstrated mechanisms: the survey recorded stated intentions, not actual rides or regulatory decisions.
Self-reported AI experience was also associated with greater support for autonomous vehicles and surgery. That does not establish that exposure causes acceptance. Experience was measured through reported use at home or work and algorithmic media recommendations, rather than an objective knowledge test. The findings also cannot automatically extend beyond U.S. local officials, and incomplete responses remain a limitation despite robustness checks.
For teams assessing AI adoption, the practical lesson is to measure policy support and willingness to use separately. Neither measure establishes safety, successful deployment or measurable value. This study shows how approval can conceal a consequential difference in readiness.
Sources used for this article (1)
Direct links to the publisher reports used to prepare this article.
- Source 1
- What influences attitudes about artificial intelligence adoption: Evidence from U.S. local officials Michael C. Horowitz, Lauren Kahn
