But the AI of it all adds a layer of opacity to these interactions. Confused driverless cars seem to make random stops<\/a>, their mandatory backup sounds are irritating<\/a>, and it\u2019s unclear whether they are even beholden to traffic laws<\/a>, human drivers allege. When it comes to parking in neighborhood spots, some residents are annoyed, and some are unsettled by the always-on cameras and sensors. But most just want to know why their homes or blocks seem to have become unofficial Waymo hubs.<\/p>\n \u201cOf all the blocks\u2026\u201d Delgin wonders. Why hers?<\/p>\n Dozens of new driverless Waymo cars on the streets in West Los Angeles. Photo by Citizen of the Planet \/ Universal Images Group via Getty Images<\/p>\n Morgan, for her part, is delighted by the Waymo\u2019s return visits, since she sees it as approval of their family by the robot cars. Her parents have been more curious, so they\u2019ve run some informal experiments.<\/p>\n They\u2019ve learned that it\u2019s not just one Waymo taking the spot, since Tucker has documented the license plates of multiple different vehicles. A Waymo will also only select one of two specific parking spots: directly in front of their home or straddling the property line with their neighbors to the south. Delgin has noticed that if both of those two spots are taken, a Waymo coming through will slow down, but not stop or try to park elsewhere on the street, even for open spots directly in front of, behind, or across the street from their home.<\/p>\n The Delgin \/ Tucker family is not alone in noticing the repeated presence of a parked Waymo in LA, where the robotaxis have operated since November 2024. A resident of the Pico \/ Fairfax neighborhood, Tal (who preferred to be identified by her first name), said a Waymo often parks on her street, always in front of an apartment building a few doors down. The Verge has observed a Waymo frequently at the same location on a street in Brentwood. Multiple users on Nextdoor have asked fellow neighbors (or complained) about the phenomenon in neighborhoods across LA, including Palms, Playa del Rey, and Westchester.<\/p>\n 1\/4Residents go on Nextdoor and other sites to complain about the loitering Waymos. Image: Nextdoor<\/p>\n The practice has also been ongoing in Arizona, where Waymo has operated since 2020. One Scottsdale family says a Waymo often parks around the corner from their home, which is adjacent to a shopping center. In recent months, Reddit<\/a> users in Phoenix have documented the robotaxis\u2019 frequent spots, while a 2023 AZCentral article<\/a> first documented the phenomenon. \u201cHow do you stop Waymo from storing cars in front of your house between trips?\u201d asks one Reddit thread<\/a>.<\/p>\n Whether a human or AI drives a taxi, it makes sense that a car would idle for some time before it\u2019s dispatched on its next ride. But what makes Lisa Delgin\u2019s home or the Pico \/ Fairfax building a safe haven for the robotaxis, with such apparent specificity? Neighbors have their theories: proximity to high-traffic areas, central locations, a lack of parking restrictions, ample curb space. But none of that accounts for the repeated specificity of the parking choice.<\/p>\n And unfortunately, not even Waymo the company necessarily fully knows the answer.<\/p>\n Can a robot catch a break?<\/p>\n Where does a Waymo go on its downtime? The company acknowledges that street parking may be a part of a Waymo\u2019s daily routine. There are parking depots throughout LA where Waymos get charged and cleaned. But when not in use, the company says that the cars park outside the Waymo lots, too.<\/p>\n \u201cOur vehicles will find appropriate parking spots to wait for short periods between trips, either in Waymo\u2019s parking facilities or on-street parking locations,\u201d Vishay Nihalani, Waymo\u2019s director of product management, said in a statement. Waymo is even participating in a forthcoming study from UC Berkeley and UC Irvine analyzing street parking behavior, and its impact on ride wait times and curb and road congestion, by providing researchers with aggregate \/ hypothetical data.<\/p>\n Waymo street parking behavior differs from a cab or other rideshare vehicle. After the morning rush, Uber driver \/ actor Josh Myra heads home, which is naturally where he parks his car when he\u2019s not taking rides. Otherwise, Myra does not make a practice of parking in residential neighborhoods between rides. When he\u2019s not working, he\u2019s parked at home. And if he is looking for rides, he will usually drive around high-traffic areas, a practice known as \u201cdeadheading,\u201d since he gets far more ride requests when he\u2019s in motion versus when he\u2019s stopped. Myra said this is typical of most of the Uber drivers he knows.<\/p>\n But if a Waymo isn\u2019t at a depot, or doesn\u2019t need to head to a depot for a charge or service, it\u2019s not necessarily going to act like an Uber driver and chase rides. If Waymo isn\u2019t seeing a high amount of traffic for ride requests, it may choose to park to conserve energy and avoid contributing to traffic, while also ensuring there\u2019s some Waymo coverage in non-high-traffic zones.<\/p>\n \u201cWhen Waymo vehicles are idle and don\u2019t have charging or maintenance needs, they choose between parking in nearby spots or driving to areas of high demand,\u201d Nihalani said. The Waymos opt to drive to these \u201chigh-demand\u201d areas when \u201cthey\u2019re likely to quickly receive the next hail.\u201d If high-demand areas have an adequate number of Waymos nearby, the cars may choose to park. \u201cThis allows us to best match ride-hailing demand and vehicle supply, while conserving energy and reducing traffic congestion,\u201d Nihalani said.<\/p>\n As for where to pull over, there are a few things that definitively go into Waymo\u2019s parking choices. First is \u201clocal parking regulations,\u201d so that includes the actual curb restrictions such as time limits or street sweeping. Next is \u201cthe number of our vehicles that are parked in a given area.\u201d If there are already Waymo vehicles servicing that neighborhood, the individual Waymo is likely to move on. And finally, \u201chow long they remain parked.\u201d If they tend to hang out there for long periods of time, that could mean a Waymo might not park in that neighborhood (or spot) in the future, since it indicates there\u2019s not much demand in that area.<\/p>\n The decision to park or head to a depot or a high-traffic zone appears to be a balancing act. If a neighborhood doesn\u2019t have another Waymo nearby, and high-traffic areas are saturated with Waymos, parking in a place with minimal curb restrictions might be the best choice. The curb traffic study supports this: UC Irvine lead researcher on the curb study Michael Hyland<\/a> said data showed that street parking reduces wait times and road congestion. Although it does, obviously, take up more parking spots.<\/p>\n Photo by Mario Tama \/ Getty Images<\/p>\n While Waymo provides general insight into its parking choices, it does not address the specificity of the places its vehicles choose to stop \u2014 in front of the same house or apartment building, over and over again.<\/p>\n Waymo acknowledges the specific repeat parking phenomenon, but says that the cars\u2019 AI is making so many dynamic decisions in driving as well as parking \u2014 about safety, accessibility, and congestion \u2014 that it can\u2019t say why a Waymo selects a spot for idling or parking.<\/p>\n Neither the repetition nor the lack of clarity around the behavior shock Phil Koopman<\/a>, a Carnegie Mellon professor and autonomous vehicle expert. \u201cA computer just doing exactly the same thing the same way every time should not be a surprise to anyone,\u201d Koopman said. \u201cThat\u2019s how computers are.\u201d Essentially, they\u2019re smart enough to find an absolutely optimized parking spot \u2014 but not necessarily elastic enough in this instance to choose a nearby spot if the optimal spot is not available, the way a human would.<\/p>\n As for how the computers are arriving at that result, Koopman has another theory. Waymo saying that only the car\u2019s computer really knows why it does what it does leads him to believe that Waymo is using machine learning to direct its curb parking decisions. Machine learning<\/a> means that Waymo engineers provide the cars with data that allows them to make their own decisions based on that data, rather than provide specific instructions via a human-written algorithm.<\/p>\n \u201cMachine learning looks at statistical information and comes up with an answer, and nobody has any idea how it got there,\u201d Koopman said. \u201cAnd it seems that they are currently not optimizing to vary the location for whatever reason.\u201d<\/p>\n Waymo does not confirm or deny its use of machine learning, instead saying \u201cseveral elements go into the vehicles\u2019 determination of where to park,\u201d according to Nihalani.<\/p>\n Koopman says that while Waymo may not have insight into how the cars are arriving at their parking choices, they absolutely understand what it\u2019s basing those decisions on. \u201cThey don\u2019t know why it made the decision, but they know what data they\u2019re feeding it,\u201d Koopman said. Where it\u2019s legal to park is certainly in this data, but one Reddit user<\/a> suspects that Waymos have \u201cdesignated safe spots\u201d \u2014 a suspicion that Koopman shares. However, Waymo does not confirm the existence of such designations, and only says that it prohibits parking in certain places.<\/p>\n Tal, the Pico \/ Fairfax resident, feels slightly uncomfortable about the constant presence of the Waymo with its always-on cameras; Redditors have expressed similar sentiments. Delgin and Tucker have been mostly amused, though sometimes annoyed by the car taking up a spot. Once, while about to head out on an errand, Delgin even made a quick U-turn back into her street spot when she saw a Waymo coming to block it. This sort of moderate annoyance is the overwhelming sentiment on forums where people raise the issue, too.<\/p>\n Some residents have taken their complaints directly to Waymo, the company says. The thing is, Waymo is not technically doing anything wrong, as long as parking doesn\u2019t exceed three hours. Regarding parking regulations governing Waymo parking, Los Angeles Department of Transportation spokesperson Colin Sweeney cites Los Angeles Municipal Code 80.69.2(b), which says that commercial passenger vehicles under 22 feet must follow the same parking regulations that personal vehicles do, and that they can\u2019t park in the same spot for over three hours.<\/p>\n Neither Tal nor Delgin is sure whether Waymo has overstayed this time limit. But researcher Hyland says that, according to the aggregate \/ hypothetical data, very rarely did any rideshare vehicle park for more than two hours.<\/p>\n Waymo can, and has, told its vehicles to steer clear of certain spots. Waymo confirmed that it is technically possible to mark a spot as a no-parking zone for Waymo vehicles, which it has done in response to neighbor complaints. The police departments of Los Angeles and Phoenix weren\u2019t able to say whether they had gotten any general \u201cloitering\u201d complaints, as these types of information requests require that inquiries be pinpointed to specific locations.<\/p>\n \u201cWe are committed to being good neighbors in the communities we operate in,\u201d Nihalani said. \u201cWe have received a few pieces of feedback from neighbors, and have made adjustments accordingly.\u201d<\/p>\n Being a \u201cgood neighbor\u201d is all well and good, but not providing clarity into the specific parking behavior could indicate larger problems if it is also using machine learning for other functions.<\/p>\n \u201c[Parking] is not high stakes,\u201d Koopman said. \u201cBut companies like to use, \u2018Well, the computer just did what it did, we don\u2019t know,\u2019 as an excuse for shirking accountability for decisions that might be somewhat harmful.\u201d<\/p>\n Lately, the Delgin \/ Tucker family\u2019s Waymo has been around less. Waymo confirms that, while usage fluctuates, it has seen a significant uptick in recent months. So chances to stop and take a breather for the robotaxis might be fewer and farther between as the supply of Waymo\u2019s 500-strong Los Angeles fleet begins to meet Angelenos\u2019 demand. Lisa Delgin just hopes, in the future, a Waymo will be around when she needs it.<\/p>\n 3 Comments<\/a>Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.Rachel KrausCloseRachel Kraus<\/p>\n Posts from this author will be added to your daily email digest and your homepage feed.<\/p>\n PlusFollow<\/p>\n See All by Rachel Kraus<\/a><\/p>\n Autonomous CarsCloseAutonomous Cars<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n PlusFollow<\/p>\n See All Autonomous Cars<\/a><\/p>\n ReportCloseReport<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n PlusFollow<\/p>\n See All Report<\/a><\/p>\n TransportationCloseTransportation<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n PlusFollow<\/p>\n![\"Dozens](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2025\/08\/gettyimages-2227598050.jpg\"\/)
<\/a><\/p>\n![\"Waymo\u2019s](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2025\/08\/2085264814.jpg\"\/)
<\/a><\/p>\n