City Detect mounts cameras on public vehicles like garbage trucks and street sweepers, captures photos of surrounding buildings as those vehicles pass, then uses computer vision to analyze the images. It’s essentially a Google Maps Street View, but focused on ensuring buildings are up to code. “The problems could be graffiti, illegal dumping, litter that’s on the side of the road,” Baum-Blake told TechCrunch. Then, City Detect works with local governments to fix the issues, a process that usually involves local officials sending a crew out to clean everything up. Right now, tracking dilapidated buildings is very manual, so Baum-Blake considers his competition to be the “status quo.” “They’re able to do 50 per week,” he said of humans tasked with keeping track of decaying buildings, “whereas we’re able to do thousands per week.” The product, which Baum-Blake has patented, has some fun and essential features. The latter is that faces and license plates are always blurred for privacy reasons; the former is that City Detect’s technology can distinguish between street art and vandalism. It also helps governments track whether landlords are not properly maintaining their buildings. “We’re able to see if there’s structural roof issues or we’re able to identify if there’s been storm damage,” Baum-Blake continued. City Detect is in at least 17 cities and works with local governments in places like Dallas and Miami. The company has raised $15 million in funding to date and is a member of the GovAI Coalition (an AI governance collective), is SOC 2 Type II compliant (meaning it’s independently certified for privacy), and follows its own responsible AI policy. “We published our Responsible AI policy in response to a consortium of local governments that stated they were looking for clarity on what vendors were actually willing to commit to,” Baum-Blake said. “We committed to this policy so that our local government partners could know what to expect from us.”