These robots feel anxious about aggressive drivers, get confused by unusual pets, and even go through a type of culture shock when relocating from the West Coast to the East Coast. A recent presentation by an autonomous delivery executive revealed that the artificial intelligence powering today’s sidewalk robots faces challenges that are surprisingly human-like.
While the public often envisions autonomous robots as cold, logical machines, deploying them in public spaces shows this technology is deeply focused on social acceptance and survival. , the co-founder and vice president of product design for , addressed the conference with the argument that .
The baby goat as a ‘long tail’ problem
Issues usually start when these machines leave the controlled environment of simulations and enter the “wild” of city sidewalks, Burk Chun explained. During a deployment in Los Angeles, the delivery team found the real world was “even more dynamic than we expected.”
In one case, a robot froze, “totally confused by the pet baby goat” in its path. While the robot’s sensors could identify human pedestrians, the goat was a “long tail problem”—a rare event standard training data hadn’t prepared the AI for. Like a person seeing something inexplicable on their commute, the robot simply didn’t know how to react.
Robotic nightmares on Main Street
Confusion isn’t the only issue for these droids; fear is too. Intersections are described as “one of the most dynamic spots in our cities,” filled with fast-moving vehicles that pose an existential threat to small delivery devices.
“Robots have nightmares about cars,” the executive said, without explaining how she detects robotic nightmares or what they entail. “Cars are also very scary for robots.”
She noted that robots must constantly calculate risks of sharing public spaces with heavy vehicles. To manage this, engineers spend significant time determining if a robot is “safe enough to cross the street,” assessing everything from pedestrian signals to ground conditions.
Cross-coast culture shock for robots
For anyone who’s relocated, one of the most relatable struggles is adjusting to local culture. It turns out robots aren’t exempt from this.
The company found that “conservative routing” algorithms optimized for Los Angeles—designed to handle “high-traffic, fast-speed intersections”—didn’t work well when the fleet expanded to Florida. In Miami Beach, drivers tend to “cruise” instead of rushing to turn like Angelenos, so the robot’s hyper-cautious LA programming was out of sync with local rhythms.
“The future is already here … it’s just not evenly distributed,” Burk Chun said, paraphrasing sci-fi legend William Gibson, who first popularized cyberspace in the 1980s. ( is a notable Gibson classic.)
“It’s also amazing how each city shows its character through how people walk,” Burk Chun added. “Not just sidewalk infrastructure, but how people drive.” She said every city has a unique “flavor” robots must learn when moving there—just like humans.
Robots as neighborhood guests
Underpinning these anxieties is a strict social contract. “Robots don’t have rights to be on sidewalks—people do,” Burk Chun asserted. This philosophy means engineering decisions must be “socially aware,” prioritizing human comfort over robotic efficiency.
Since “more people will walk next to the robot than receive a delivery from it,” the machine is viewed as an ambassador. If it fails to “deliver delight” or add value to the community, it’s seen as a missed chance to build a harmonious future.
To earn their place, these robots do more than deliver meals; they act as municipal inspectors. Using advanced sensors, they collect data on “missing curb cutouts” and “hidden potholes,” sharing this information with cities to help repair infrastructure.
For this story, journalists used generative AI as a research tool. An editor verified the accuracy of the information before publishing.