On June 23, 2026, TaskUs (Nasdaq: TASK) released its 2026 Autonomous Vehicle Trust Consumer Survey, providing the most comprehensive data snapshot yet of American consumer attitudes toward robotaxis. The survey of over 1,000 U.S. consumers, conducted in May 2026, yielded a headline finding—38% of Americans are likely to try a robotaxi within the next 12 months—that positions the autonomous vehicle industry at the threshold of the early majority market. But the numbers themselves are less interesting than the structural pattern they reveal: trust is not built through publicity, but through exposure.
Core Data: The Curiosity-to-Trust Ladder
The survey reveals American attitudes toward robotaxis at a fascinating equilibrium: 42% of respondents reported being “curious,” while 41% felt “nervous.” The single percentage point gap between curiosity and nervousness precisely captures autonomous driving technology’s current social-psychological position—neither enthusiastically embraced nor vehemently rejected, but suspended in a “waiting to be convinced” equilibrium.
More illuminating are the contrasts between robotaxi-operating cities and non-operating cities. In cities with existing robotaxi services, 37% of residents have already ridden one; in non-robotaxi cities, that figure is just 6%. Even more striking is the difference in “full trust”: 23% of residents in robotaxi cities fully trust autonomous driving, compared to only 4% in non-robotaxi cities.
TaskUs Head of AV Nick Allen’s summary of this finding is incisive: “The biggest barrier to robotaxi adoption is not safety concerns, but a lack of access.” This runs counter to many industry intuitions—the prevailing assumption has been that safety records are the primary obstacle to autonomous driving adoption. The data suggests otherwise: once people have a single real experience riding a robotaxi, their trust levels increase dramatically.
The generational divide tells an equally compelling story. Among Gen Z (ages 18-26), 29% fully trust robotaxis; among Baby Boomers (60+), just 11%. Only 13% of Gen Z say they are “very unlikely” to try a robotaxi, compared to 48% of Baby Boomers. This gradient reveals an important structural dynamic: the cohort that has grown up with autonomous driving technology is naturally aging into the market.
Trust Data vs. Operational Reality
TaskUs’s survey findings align closely with actual operational data from major robotaxi operators. Waymo—currently the largest U.S. robotaxi operator—reached over 1 million paid weekly trips in 2025, a number that has continued growing in 2026. Its service operates across San Francisco, Los Angeles, Phoenix, Austin, and is expanding into Miami, Atlanta, and other new markets.
Zoox (Amazon-owned) expanded its Las Vegas operational area in 2026, accumulating increasing public road experience with its unique bidirectional, steering-wheel-less design. Pony.ai continues advancing in both China and the U.S., targeting 3,000 robotaxis across 20+ cities by end of 2026.
Notably, Tesla’s robotaxi fleet still numbers just 59 active vehicles—a stark contrast to Elon Musk’s ambitious promises. While Tesla’s FSD system continues improving (v13 fully adopts end-to-end neural networks for unprotected turns and roundabout navigation), the gap between driver assistance and true driverless robotaxi operation is proving far larger than many expected. This indirectly validates another TaskUs finding: consumers’ trust in “autonomous driving” is not uniform—people distinguish between “driver assistance” (Tesla FSD) and “driverless robotaxis” (Waymo) more effectively than the industry may have anticipated.
From “Safety Is Relative” to “Safety Is Absolute”
Thirty-five percent of respondents believe robotaxis are safer than human drivers, while 48% require proven safety data before their trust would increase. This pair of numbers reveals a structural communication dilemma for the autonomous driving industry. Despite Waymo and other operators accumulating tens of millions of real-world autonomous miles with collision rates far below human drivers—Waymo’s robotaxis have approximately 5x lower collision rates than human drivers—public expectations for “absolute safety” far exceed “relative safety” comparisons.
This asymmetric expectation is a classic psychological phenomenon known as “algorithm aversion”: when machines make mistakes, people’s tolerance is far lower than when humans make equivalent errors. The autonomous driving industry faces a challenge that extends beyond technology to a deep social-psychological question: will humans accept a driving system that, while 5x safer than human drivers, is still not perfect?
Policy and Infrastructure Alignment
Trust is not solely a function of technology and data; policy context plays a critical role. The U.S. federal government has not yet passed comprehensive autonomous driving legislation. NHTSA continues regulating through administrative guidance and exemptions, lacking a unified federal framework. State-level regulatory divergence creates operational uncertainty: an autonomous system approved in Arizona may require an entirely new approval process to operate in New York.
In China, the regulatory framework is comparatively more unified. The Ministry of Transport issued a series of autonomous driving management regulations between 2025 and 2026, clarifying approval processes for testing, demonstration, and commercial operation. Pony.ai, Baidu Apollo, and DiDi autonomous driving have obtained operating permits across multiple cities simultaneously. However, China’s complexity lies in its extremely dense urban traffic environments—autonomous driving on Guangzhou or Beijing’s congested streets presents far greater technical challenges than Phoenix’s wide desert roads.
Infrastructure alignment matters equally. Cities with robust V2X (vehicle-to-everything) infrastructure—traffic signal communications, roadside sensing units, high-definition maps—enable significantly safer robotaxi operations. But building this infrastructure requires massive investment, and the return depends on autonomous driving penetration rates—a classic chicken-and-egg problem.
Outlook: From Pilot Cities to Mass Adoption
The most optimistic reading of the TaskUs survey: 38% willingness-to-try implies approximately 100 million potential robotaxi users in the United States alone. But the distance from “willing to try” to “regular user” remains enormous. Current robotaxi per-mile costs—though declining rapidly—still exceed the marginal cost of private vehicle ownership, especially when purchase costs are disregarded.
Looking 3-5 years ahead, robotaxi adoption will follow a clear S-curve: natural growth in cities where service already exists (trust built through exposure), acceleration as more cities open for operations, and finally reaching critical mass among mainstream consumers. The ALOT (Autonomous Last One Truster)—the last person to trust autonomous driving—may not appear until approximately 2035. But the industry’s inflection point likely arrives in 2028-2030, when robotaxi operating costs fall below human-driven Uber/Lyft and major city regulatory barriers are substantially cleared.
Disclaimer: The information in this article is for reference only and does not constitute investment advice or business decision-making basis. Data and time information are current as of the publication date and may change with subsequent developments. Neither the author nor POC.HK assumes any responsibility for losses resulting from the use of this information.