Robotaxis are failing to recognize and yield to emergency vehicles, leading to dangerous delays during critical moments. A fatal gas explosion in Dallas in late May exposed a critical gap in autonomous vehicle safety when a Waymo robotaxi blocked firefighters' access for over three minutes. Another incident in Austin saw a Waymo freeze across the path to an active crime scene, forcing a police officer to manually move the vehicle. These failures have prompted Texas to introduce formal licensing requirements, mandatory emergency plans, and stricter regulatory oversight for autonomous vehicle companies operating in the state.

What Safety Issues Are Robotaxis Creating on Roads?

The problems extend far beyond emergency response delays. CNN identified hundreds of incidents in which Waymo robotaxis made dangerous moves that human drivers would instinctively avoid. These include running red lights, driving into oncoming traffic, entering active crime scenes, failing to observe emergency road closures, and coming within inches of cyclists and pedestrians. In just the past two months, Waymo has recalled thousands of vehicles and halted operations in multiple cities after robotaxis drove into flooded streets, including an incident in San Antonio where an unoccupied Waymo was swept away in rushing floodwaters.

The incidents reveal a fundamental disconnect between how autonomous systems perceive and respond to real-world emergencies. Robotaxis make decisions based on sensor technology, object recognition, and route logic, but they struggle with boundary conditions that humans navigate instinctively. When a robotaxi can remotely unlock doors but only reacts after official identifiers are provided, people could be trapped inside for too long during an emergency. Similarly, robotaxis getting stuck while maneuvering shows that even with strong driving statistics, certain situations cannot be sufficiently fail-safe in emergency mode.

How Are Regulators and Companies Responding to These Failures?

• Texas Regulations: The state is imposing stricter rules on autonomous vehicle companies, including formal licensing requirements, mandatory emergency plans, public complaint services, and stronger regulatory intervention powers to address safety gaps.
• Waymo's Product Evolution: On June 3, Waymo launched a new robotaxi model called "Ojai" in several cities, developed in collaboration with Zeekr and powered by a sixth-generation version of Waymo's software, though it remains unclear whether these changes will improve emergency response capabilities.
• Development Priorities Shift: Companies are increasingly aligning their AI development and infrastructure with compliance requirements, signaling that regulatory pressure is reshaping how autonomous vehicle technology is designed and deployed.

Waymo maintains that its robotaxis have already made roads safer overall. The company claims its vehicles are 13 times less likely to be involved in serious injury crashes compared to human drivers. However, this statistical advantage does not address the specific failure modes that have emerged in emergency situations, where robotaxis must interact with first responders and navigate unpredictable, high-stakes scenarios.

The incidents in Texas are not isolated. In Atlanta this year, dozens of empty, driverless Waymo vehicles swarmed a residential neighborhood, repeatedly circling and clogging a quiet cul-de-sac after a software glitch routed multiple passenger-less robotaxis into the same dead-end loop, leaving locals frustrated and concerned for the safety of children and pets. In China, where thousands of robotaxis are in operation, similar problems have surfaced. When more than 100 robotaxis in Wuhan came to a standstill, operator Baidu provided only a vague explanation of a "system failure" and did not respond to media inquiries, raising concerns about transparency and accountability.

Safety experts and industry observers agree that early-phase autonomous vehicle failures often stem not from perception alone, but from a disconnect between autonomous decision-making and the interpretation of external, human signals. Traditional driver assistance systems are designed for defined driver interaction, but robotaxi operations require a comprehensive emergency concept that accounts for first responders, pedestrians, and unpredictable road conditions. The key question now is whether the technology can keep pace with the new rules being introduced across jurisdictions.

Even those calling for caution support the long-term potential of the technology, saying it will improve and has the potential to transform road safety, but only if issues that surface are taken seriously by both companies and government regulators. The incidents in Texas serve as a litmus test for authorities worldwide, who must ensure not just that robotaxis can drive, but that they can integrate seamlessly into unpredictable real-world scenarios where human lives depend on split-second decisions.