Podcast’s Essential Bites:
[3:26] ES: "The challenge is we don't actually know where all the lead pipes are. [...] What we do know is largely based on these outdated records, which are sometimes hand drawn maps and books of atlases, or on index cards and file cabinets in the basement of City Halls [...] or maybe they're actually now digitized in GIS databases, but they're based on those. Those don't necessarily reflect the reality underground [...] and often they're incomplete."
[5:39] ES: "We were having conversations with community members and officials in Flint in 2016. And we realized that to answer a simple question of how many lead pipes are there in Flint after the international news media came and left was still very difficult. And it's because of the uncertainty caused by all of these old records and lack of record keeping for decades, meant that we just didn't know."
[6:22] ES: "We resolve that uncertainty by finding patterns that are connected with homes that likely have lead and then we extrapolate and we predict for all the other homes for which we've not yet verified. [...] What's the likelihood that each of those homes has lead based on the age of the home, the location of the home, the size of the lot, the condition of the all the other variables that we actually do have, based on tax records about the parcel of land, or existing water tests? Based on all of that we're able to pattern match and start assigning probabilities."
[8:22] ES: "Sometimes [people] just say anything built after 1986 has no lead, anything before 1986 we assume has lead. That is when the federal government banned the use of lead, but that's an imperfect rule. [...] What AI is doing is, it's saying, it's not just one variable. And it's not just a simple relationship, like older means more lead, but that there are lots of different variables involved. And it's the combination of them that's going to let us start getting better assessments."
[12:55] ES: "We're just at the point where we do have enough data from a variety of cities across 10 states, 50+ cities, where we are now making predictions at the aggregate level for a whole water system. Even though we have not worked with a client in that water system, we now have estimates for every water system in the country. How many lead service lines do we expect them to have right now? [...] And that is where I think there's a lot of power to some of these predictions to inform the way the states can start allocating their State Revolving Funds that are coming from the Infrastructure and Investment Jobs Act."
[15:51] ES: "The Rockefeller Foundation has just granted us $1 million to scale BlueConduit's artificial intelligence and machine learning centered approach to identifying and removing lead pipes. [...] It's specifically helping us grow our free and open source software [...] and building out on top of all that a nationwide interactive map."