Monday 8/17/26: Capitol Notes, newsrooms and AI, The Great Oconto Frog Invasion

David Lee: Support for WUWM podcasts comes from Potawatomi Casino Hotel Milwaukee, home to 24-7 excitement, dining, and an on-site hotel at 1721 West Canal Street. More information at potawatomi.com

Joy Powers: From 89.7 WUWM, Milwaukee's NPR, this is Lake Effect. I'm Joy Powers. Today in a new Capitol Notes, we'll examine the race for governor and what's ahead in the general election.

JR Ross: With Crowley as the nominee now for Democrats, it makes it easier for Dems to prosecute the case against Tiffany via gas prices and Donald Trump than if it was Hong that Tiffany was running against.

Joy Powers: Then we'll explore how local newsrooms are grappling with AI, and look at ways it can be used as a tool.

Alex Mahadevan: This is not about using a tool to do the stuff you love to do. This is using a tool to make sure you have more time to do the things that you love to do in your job.

Joy Powers: Plus, we'll tell you about the 1952 Oconto Frog Invasion. All of that is coming up on Lake Effect. This is Lake Effect, from 89.7 WUWM, Milwaukee's NPR. I'm Joy Powers, and thank you so much for joining us. We have now entered general election season in Wisconsin. Last week, Milwaukee County Executive David Crowley defeated State Representative Francesca Hong for the Democratic nomination for governor. Crowley will face U.S. Representative Tom Tiffany, who handily won the Republican nomination. Here's a new Capital Notes, looking at what's ahead in the race for governor.

Maayan Silver: This is Capital Notes from Milwaukee's NPR. I'm Maayan Silver, speaking with JR Ross, editor of wispolitics.com, in a special post-primary edition. So, hi there, JR. Thanks for joining me.

JR Ross: Oh, thanks for having me.

Maayan Silver: What a wild election night. It was teeter-tottering between tenths of points, sometimes a dead heat, in the Democratic primary race between Milwaukee County Executive David Crowley and State Representative Francesca Hong. What are your big takeaways from the night?

JR Ross: A couple of things. One, there was obviously a surge for Crowley in the end. Now, there's a debate about whether the polling ahead of the primary was any good, if it missed what was going on. But what is clear is that Crowley came from dropping out of the race July 8th, getting back in July 18th, to being the nominee on the early hours of August 12th. So it's quite a comeback. And there's combination of things: one, give credit to Crowley and those around him. They spent some $4 million at least on paid media. Governor Evers was a big assist, campaigning with Crowley in the end. And the third thing was that Francesca Hong kind of dropped the ball a bit in the closing days of the primary that hurt her. And, talking to insiders, one of the things they noted was that, you know, after Mandela Barnes dropped out July 30th, rather than kind of saying, "Okay, I'm in the driver's seat here; I'll try to pull back," Francesca Hong went on national cable news and talked about Thanksgiving and how she had one time tweeted about calling for it to be canceled. Like, these are not things you do if you are kind of a seasoned candidate or have a seasoned staff around you. She made some real mistakes in how she interacted with the media. And if you add that in with people starting to say, "Hey, she can't win in November," and some other factors, that probably helped fuel that Crowley surge in the end.

Maayan Silver: Okay. So looking forward, Crowley ended up winning the Democratic primary, while Republican Congressman, Tom Tiffany, ended up winning the GOP primary for governor. Looking forward, can you give a short synopsis of those two candidates strengths and weaknesses?

JR Ross: Well look, Crowley does not have the baggage that Hong did of tweeting about canceling Thanksgiving, or being uncomfortable at a Culver's in northern Wisconsin because she thought there were a lot of white people there who assumed that she didn't speak English, stuff like that. He has a lot the same kind of beliefs in terms of policies, but Crowley is making the case that he's gotten stuff done. He's been county executive. Well, Republicans, we're already seeing, are going to attack him on things like raising taxes in Milwaukee County, crime in Milwaukee County, the issues with the health insurance plan for county employees, that there was a little lapse there and how it was handled. They're going to make the case that he's just as liberal as Hong, but just does things a different way. For Tiffany, it is that he is basically tied at the hip to Donald Trump. He is staunchly anti-abortion. And he's part of Washington, D.C. You know, I mean, obviously, Tiffany represented the 7th Congressional District, served in the state legislature, but he's a part of that D.C. world, because he serves in Congress. That's not a popular place to be. And in the end, you know this race is going to be shaped a lot by a couple of factors. Nationally, thinking about, one, just the cost of living and especially the cost of gas. And two, Donald Trump, what are his numbers? And people have argued to me in the past few days that, with Hong, it would have taken the focus off of Trump and the economy and put it more onto Hong and her beliefs and her past social media comments. With Crowley as the nominee now for Democrats, it makes it easier for Dems to prosecute the case against Tiffany, via gas prices and Donald Trump than if it was Hong that Tiffany was running against.

Maayan Silver: It seemed like no one was expecting that primary, that Democratic primary result. Do you think Republicans have been caught flat-footed on that as well, like expecting specific ads against Hong that they already paid for, that they can't use, for instance?

JR Ross: Well, there was definitely a pivot point. They had to pivot. I mean, they were expecting Hong to get in. And you actually saw it in the week before the primary where Hong became this like national talking point in the conservative media ecosystem of, "Oh, look how out there she is. She's the new face of Democrats. She's going to be their nominee for governor. She's got all these problems. We're going to start going after her." And in fact, the Republican Governors Association has a thing called Right Direction Wisconsin PAC, which is a group that is actually, you know... They're a spending vehicle in Wisconsin gubernatorial races. They began an ad by mid-July that was supposedly kind of an opposing Hong, but really it was designed to kind of boost her in the primary. It was talking about she opposed Donald Trump's policies on immigration and ICE. It featured a clip of her saying in an interview that she was the only candidate in the primary who was going to take the Democratic Party back to its progressive roots. So it's like a wink and a nod kind of thing, like, yeah, it's quote-unquote "opposed" to Hong, but there were really trying to support her and wanted her to win the primary. Well, July 31st, he switched traffic and folk and brought in a new ad in that was more of a contrast with Tiffany. It still highlighted how Hong opposes Trump's immigration policies, but he talked more about how she wants to raise taxes. And it's just an example of where Republicans began to count on the idea that she was even the nominee, and they were gonna treat her as such. And it probably didn't help them in the end that they began kind of this barrage of attacks on her, because they wanted to find her early. And it ended up probably helping probably a little bit

Maayan Silver: This is Capital Notes. I'm Mayan Silver speaking with J.R. Ross of wispolitics.com. You've talked about how important it is for national Democrats, or important donors to the Democratic Party, to get behind the Democratic nominee in Wisconsin for them to be able to kind of have some legs against Congressman Tom Tiffany. Are you seeing that yet? Is it too early to tell? How is that going?

JR Ross: We've already got a buy from the Democrat Governors Association going after Tiffany. It launched the day after the primary. It's about a $1.4 million that I've seen so far. So yeah, they're, they are going full tilt. And that was a big question was whether they support Hong. But now with Crowley, they're definitely getting on board in this race.

Maayan Silver: Does this new head-to-head match-up between Crowley and Tiffany, how will it impact the efforts of both parties: in Republicans' case, to keep control of the Senate and Assembly, and in Democrats' case, their effort to flip them?

JR Ross: Look, a lot of the seats that are up for grabs this fall, Kamala Harris did better than Donald Trump in 2024, which was not a great year for Democrats. There was a real fear among Democrats and among Republicans that Hong as the nominee would make it easier to win those seats, because she would be a drag on the Democrat brand up and down the ticket. Now with Crowley, it's a much more back in Democrats' favor in the State Senate where nothing is guaranteed, but they feel much more confident about the Senate now. Republicans, they have a tougher fight on their hands with Crowley as nominee. Look, Crowley still has baggage. He's still gonna get knocked about a bunch of things, but he doesn't have the baggage that Hong does. It changes the equation quite a bit. We're also now gonna see, again, where's Donald Trump come October for his job approval ratings? How expensive is gas? You know, a lot of times, the wins coming out of Washington, D.C., shape elections because of what's going on. The winds right now are blowing against Republicans because Donald Trump is not very popular and gas prices are high. If that changes in the next two months, different ballgame. If it stays that way, that makes it that much more difficult for Republicans, because again, the national mood shapes even local elections so much that it's going to be hard for them to escape that shadow of the president.

Maayan Silver: And as a congressman, Tom Tiffany, back in 2020, he voted to decertify the election results in several swing states and was going to vote that way in Wisconsin. He has been cast as an election-denier. How could the Crowley campaign use that to their favor?

JR Ross: Look, my impression is that there are certain Democrats who get really inanimate about that, but most swing voters are kind of over it. They're not paying attention to that issue that much anymore. Now, maybe that's not the right thing to take, but hey, we just know from the polls that that doesn't drive people. What drives people is their pocketbook: how are they doing financially? Are they doing okay at home? That's what's going to drive this race more than that kind of stuff.

Maayan Silver: And in a midterm year, who are those actually undecided and independent voters in Wisconsin who are still actually choosing between Crowley and Tiffany?

JR Ross: Good question. And who's going to turn out? I mean, you know, there's also, let's be honest. Crowley probably has some work to do to woo Francesca Hong supporters. Some of them are non-traditional primary voters. At least that's the impression that we got, that they were new people who were attracted to her populist message on the economy. Does, does David Crowley have any attraction to those folks, or do they feel like, "Hey, the democratic establishment really stole this from us, or kind of intervened to beat Fran. And so we don't feel really good about voting for a Democrat, period. This fall we're just gonna stay home." That's a big unknown.

Maayan Silver: What kind of coalition do you think Crowley could cobble together for himself in November, and what kind of coalition is Tiffany going after?

JR Ross: Look, Tom Tiffany needs the MAGA base to be fired up and animated to turn out, right, that he just follows what Donald Trump did. Donald Trump laid out the math for any Republican to win statewide in Wisconsin: maximize turnout in rural areas, and keep your losses down in Dane and Milwaukee counties. You don't have to win Dane County as a Republican, but you can't lose with like 20% of the vote or 18% of vote. You've got to get the low 20s. That's what Trump did. Same in Milwaukee County. Trump lost Milwaukee County, but didn't lose as badly as what Tim Michaels did four years ago. There's your math. For Crowley, follow the math that Tony Evers laid out in 2022: maximize turnout in Dane and Milwaukee counties, get the biggest margins you can. Try and hold down Tom Tiffany in Waukesha County, which is a key driver for Republicans. And then try and do better out of state than what Kamala Harris did. Rural Wisconsin has plenty of voters. You just have to kind of not lose it as much as what Kamal Harris did.

Maayan Silver: Is it possible for, or is it more of an uphill battle, when there's a competitive primary, versus, like, Republicans consolidated their support around Tiffany pretty early. I'm assuming that's a factor going in their favor, but when there is competitive primary on one side, have candidates gone on to win the actual general election even with like 40% of the primary vote, like in this year's Democratic primary?

JR Ross: Tony Evers did in 2018. He was in the low 40s. There was a 10-candidate primary, down to eight candidates who were active by August of 18. He was in the low 40s, and he won. So, yeah, it happens a lot. Jim Dole in 2002, three-way primary for governor. He won that, then went on to win a three-way governor's race, libertarian candidate on the ballot in Ed Thompson. So, yes. Seen it plenty of times. But again, it's a matter of, for Crowley, can he consolidate those voters and try and bring some of the Hong supporters onto his team.

Maayan Silver: And lastly, about election process, there was a little bit of human error on Tuesday night, the primary night, in getting the City of Milwaukee flash drives to the county. So that helped stretch out election night until about 2:30 AM when results came in. Can you explain the latest on the conversation around counting absentee ballots before election day, and how central count works in certain locations? And is there gonna be any momentum to change that before November? Is that impossible?

JR Ross: One, the legislature is not in session. We're not going to change election laws, at this point. Two, even if you had... It's not early count, it's early processing. Remember that. You could open the absentee ballots You're going to process them. You can't actually count them until election day. This wouldn't have fixed Milwaukee issue. The Milwaukee issue was pure human error, where you downloaded the results from the machines and take them over the county to like put them on the website. They didn't do it properly. They made a mistake in the first attempt. It delayed things. So that's a problem that goes beyond being able to process the ballots the day before. Republicans talked about having some kind of a hearing to bring people and talk about it. And the issue is: you can explain what happened in Milwaukee with central count, which... A central account location is where a community takes all its absentee ballots and counts them in one spot, but then takes them out to polling places to them done by the workers at those polling sites. You can tell people it was a human error, that we have a paper trail in Wisconsin, that those ballots can be counted. But I've already seen numerous conspiracy theories on social media of, "Oh, this is Democrats who, quote-unquote, 'Found the votes to stop Hong. This is why she lost, not because David Crowley had a surge in support, but because they rigged things somehow with these votes.'" Again, the ballots are currently in bags. If you want to recount, you go back and look through every ballot cast in Milwaukee County and check to make sure how it happened. But this is what happens when you have these kind of human errors: it gives fodder for conspiracy theorists and drives these things on social media that, "Oh, there's something amiss here," when it's just plain human error. And if this isn't fixed by November, it's going to be worse, especially if, let's say, it's like 2020. Tom Tiffany is down, or is leading at some point. Or, it's 50-50, and then all of the sudden Milwaukee's absentee ballots come in, and all of the sudden Crowley's up with that lead. It'll be the same kind of situation, where there'll be this conspiracy theory because of the issues Milwaukee's Central Count has had before, people will cling onto those and claim fraud, even if there is none.

Maayan Silver: What would you suggest that people do about this?

JR Ross: I've talked to numerous people who said they need to clean up their act in Milwaukee. Like, when it's human error, there's nothing more to do than get people who know what they're doing to run the show. That is the feedback I've gotten from people, and numerous phone calls, that there needs to be some kind of accountability here and some kind process put in place to not let this happen again. How did this happen, and how do you prevent it from happening again? Because, again, when you do this, it gives fodder to these conspiracy theorists who then go run with any tale they want to about what's happening in elections in Wisconsin.

Maayan Silver: All right, thanks for the breakdown, JR. And as always, thanks for joining me on Capital Notes.

JR Ross: Anytime.

Joy Powers: Artificial intelligence is reshaping how people do their jobs, and that includes journalists. Everything you hear on WUWM is written, produced, and voiced by human journalists. But there are ways it can be used as a tool in the work we do. Alex Mahadevan is the Director of MediaWise and on faculty at the Poynter Institute, where he also leads Poyntter's AI Innovation Lab. Part of his job is to research ways AI is being used to inform and misinform people. He also trains journalists to understand AI tools and use them ethically to build audience trust. He joins Lake Effect's Xcaret Nuñez to talk about how journalists are grappling with AI.

Xcaret Nuñez: A big part of your job is training newsrooms and journalists on how to use AI ethically and responsibly. What are some other concerns or reasons journalists you've talked to are hesitant about trying AI tools in their work?

Alex Mahadevan: Journalists are generally skeptical of AI tools, because I think, as reporters, we are generally skeptical of Big Tech. And these are tools that have come out of Silicon Valley and are closely associated with some of the same folks who brought us the social platforms that completely changed how we reach audiences and have pulled the rug out from under journalism before. So there is a hesitancy there just because this is a new technology put out there by, sometimes, people that we can't really trust. The second thing is how this technology was developed. These large language models were trained on our reporting. I mean, some of my reporting was scraped from the internet and put into these tools. Now, Poynter doesn't charge for its content, but for news organizations that do, I think they would see that as theft. So I think there is that concern about using a tool that was trained on our own content that we worked really hard on. And then lastly, there is definitely a concern about the environmental impact. These tools use a lot of compute, they use a lotta energy, which puts pressure on our cloud infrastructure, which has led to the development of data centers, which obviously have become a huge story for local residents who are starting to see these pop up in their backyards. So I totally see why reporters are questioning these tools, even though it is my job to make sure they use them responsibly and see a lot of opportunity in using these tools to do better work and reach more people.

Xcaret Nuñez: So before we dive any deeper, I was wondering, for our listeners, if you could explain what it means to use AI ethically in the context of journalism. I know you talked a little bit about it, but how would you describe that?

Alex Mahadevan: See, I would say using AI ethically just means using it in a way that aligns with your organization's existing editorial policies and your own moral and ethical guidelines. I do think every news organization needs a separate AI policy that generally mirrors what is in their own editorial policy and is built around principles that align with our own journalistic principles. So the other reason that journalists are very skeptical is because Big Tech does not share our values. Our values are around accuracy. They are around service. They're around accountability. They around overcoming bias and making sure all voices are heard. Now, when you think about Big Tech's values, a lot of these tools are opaque, so they are not transparent or accountable. There is not a lot thought about the bias in some of these machines that are being generated. Using these tools ethically means taking this technology and using them within our own journalistic guidelines and values. And so that means thinking about accuracy. These tools produce errors, because they are probability machines. So when you use them to do a data analysis, double checking it and making sure it's accurate. If you are using it to do something that would have made a story not possible, so let's say sifting through lots and lots of campaign finance documents and you find some really cool stories to write out of that and you write them up, you would disclose that, you know? "This tool was used to do this analysis." "A human reviewed this and this is why we did it," so you are transparent about your AI use. And, you know, overall, using it in a way that's efficient. So I teach journalists how to use these tools the least amount possible. So, like, if you use these tools really effectively, then you also are reducing your the load that you are putting on the cloud infrastructure.

Xcaret Nuñez: That's so interesting. So when you're helping newsrooms figure out what AI tools might work for them or might be best for them, I'm wondering how the audience fits into the conversation. Like, how can we as journalists preserve our audience's trust, while also using these tools to bolster our reporting?

Alex Mahadevan: Disclosures are one way. So, audiences are obviously very skeptical of AI, as journalists are. I mean, because we are, I mean I am, a citizen of a community. And so when I hear a story about a data center popping up, it does concern me. So I am also critical of AI. I teach AI, but I consider myself an AI skeptic. So I try to approach that knowing that audiences are skeptical of AI. So when we use it, we need to use it in ways that are improving their lives and that are transparent. People appreciate being told when you're using AI, why you're doing it, and how it's making their life better. The second is AI literacy videos, AI literacy columns, figuring out a way to connect with your audience to explain to them how a data center works, or how something like ChatGPT works, or if your local government has started using an AI pipeline to answer phone calls, writing about that and explaining how AI works. A study just came out from Trust In News and an academic named Ben Toff about how, if, you are in a newsroom and using AI. And if you can create AI literacy content, whether that's Instagram posts or reels or something, you do maintain that audience trust. So it's about using AI in a way that wouldn't embarrass you if people found out you were using it in this way.

Xcaret Nuñez: I'm Lake Effect's Xcaret Nuñez, speaking with Alex Mahadovan about the different ways AI is helping journalism, as well as its limitations. So to help listeners understand the way AI can actually help newsrooms, can you share an example of a newsroom that's testing AI tools successfully?

Alex Mahadevan: So, my hometown newspaper, The Tampa Bay Times, uses AI for weather alerts and for real estate briefs. And these are two pretty high-demand article types that people need in our area. And they're something that took a while to put together and took journalists off of bigger stories. But they are important, and we recognize they're important. So now, they have automated those two thing, so journalists, instead of spending so much time writing real estate briefs, can focus on bigger picture stories, bigger investigations on real estate. I mean, the Tamba Bay Times did a great investigation on hedge funds buying up homes in my neighborhood and other neighborhoods. And investigations like that would not be possible if journalists had to do some of the rote and boring things they had to do in the past. It's more time to do investigations.

Xcaret Nuñez: Yeah, so to go off that note, can you also share an example of where maybe this isn't a way newsrooms should be using AI tools or a way they can improve?

Alex Mahadevan: Yeah, so, I mean, I hate to put the Post on blast here, but the Washington Post, they've been very innovative with AI, which I respect, but they've done things like customized podcasts, which have introduced errors. This is a very trusted brand. And if someone is listening to an AI podcast that has an error in it, that is a huge problem. Also, I've listened to the podcasts, and there's sort of an uncanny valley aspect to it. So they're not, I don't think it's like the greatest product, so to speak. If you read a Washington Post story, you're sort of prompted to talk to AI throughout the story. I think that could be a pretty cool idea, but if you actually read some of the answers, they're not super helpful. So I think there is a lot to be said for experimenting with audience-facing AI. But I still think a lot of the good work is happening behind the scenes in investigative reporting.

Xcaret Nuñez: So I wanna bring you back to like more of a local lens. So say you're a local newsroom using a new AI tool and wanna be transparent about that. How do you go about disclosing how AI tools were used in the reporting process to your audience?

Alex Mahadevan: I think... So when the Tampa Bay Times did it, their top editor wrote a column about it, and said, "Here's what we're doing, here's why we're doing it, here's how we're doing it, and here's how humans are involved in the process." So it's a good way to connect with your audience just to sort of check in with them. We should be talking to our audiences more. So I would say introducing the tool itself to the audiences and doing that with column or like a vertical video or something is good. Anytime you use the tool that allows you to do reporting that you could not have done without the tool, you disclose. And in a good disclosure, you talk about what the tool is, what it did, the limitations of the tool, why you use a tool, where a human was involved. And I've seen the power of these tools, and so I would highly encourage experimentation responsibly.

Xcaret Nuñez: Yeah, well, before I let you go, I was basically going to ask you that, to build on that. It seems like you believe newsrooms have to experiment with using AI tools to survive. Or is it possible for a newsroom to stick to its ways?

Alex Mahadevan: There are very specific brands that can absolutely flourish and do great without using AI. But I think traditional news outlets who want to do really good investigative work and great data journalism are doing a huge disservice to themselves and, quite honestly, their audiences, if they are actively avoiding using something that can find secrets in campaign finance records, that can dig through a thousand emails from your county commissioners and figure out when they've been consorting with developers on a new data center. You know, there can be great reporting on data centers that can happen a lot quicker and more in depth with using these AI tools. So, I'm very passionate about traditional news organizations experimenting, but there are definitely brands who are better off not doing it and have no use for it. It is just a tool after all.

Xcaret Nuñez: Yeah, I totally appreciate that. And it's really interesting too, like, as important as it is to disclose when you use AI or how you use a AI tools, also disclosing if you don't use AI and saying that to your audience too.

Alex Mahadevan: There is definitely some value. I do think there's more value in investing and connecting with audiences via like a vertical video or community events or something than getting all hung up on proving every article is written by a human, so.

Xcaret Nuñez: Alex, thank you so much for sharing your work with me. I really appreciate your time and helping me make sense of AI in journalism.

Alex Mahadevan: And thank you. Thank you for a great conversation.

Joy Powers: Alex Mahadevan is the director of MediaWise, and he's on faculty at the Poynter Institute, a nonprofit that trains journalists and the public on media literacy and journalism ethics. He spoke with Lake Effect's Xcaret Nunez. In 1952, millions of frogs took over a city in Wisconsin. We'll tell you all about that in about 10 minutes. But first, we'll learn how local municipalities are using AI to field calls. That's coming up on Lake Effect on 89.7 WUWM, Milwaukee's NPR. You're listening to Lake Effect on 89.7 WUWM. I'm Joy Powers. If you call your city government with a request for services, you could soon be talking to AI instead of a person. A pilot artificial intelligence program has already launched in Waukesha and La Crosse counties. According to journalist Adam Rogan, it's likely to spread to other parts of the state. But what could that mean for residents in need of help, or people dealing with an emergency? Rogan joins me to explain how the AI is being used. So let's say I need to call 911. What's gonna happen next?

Adam Rogan: So right now, as it has been for past however many decades, when you call 911, you are going to get a human being operator responding to your emergency call. That is wonderfully unchanged, as human beings are still better than computers at responding to an emergency, especially when it comes to interacting with somebody who might be panicked or who might not be making sense because of the emergency going on.

Joy Powers: So good news. If there is an emergency, you call 9-1-1, you are going to be dealing with a person who is trained to deal with an emergency. So how is this AI being implemented?

Adam Rogan: So right now, this AI called Ava, made by a company called Aurelian, is in most of the state of Washington, and it's starting to spread elsewhere, including La Crosse and Waukesha counties here in Wisconsin. And what has been happening for the past, really, 20, 30 years is: emergency dispatch centers have become the redheaded stepchild of public safety. They've become the dumping ground for kind of the random calls after hours, or when a courthouse is on lunch break. If you were to call a courthouse and say, "Hey, what time is so-and-so court? Or what time my hearing?" Those calls are being forwarded to 911 operators. And a lot of times, your call is going to be put on hold because somebody's dealing with an emergency. Or these trained 911 operators are going to figure out how to help you when their job and what they want to do is to respond to emergencies. And so what a ruling does is it's this AI-trained... It's weird to call it a program, because that's not really how AIs work. But it is an AI that will answer those forwarded calls that otherwise were clogging these dispatch centers and can answer them. They'll be trained on the local database. They'll know what court hearing is tomorrow, what the current parking laws are, that kind of thing, for your local area, and they can respond to you. And the goal is to not overwork these already overworked 9-1-1 operators. The impetus for this was really realizing that the study, I think it was in 2023, that found that one third of all emergency dispatch centers in the US were more than 30% understaffed. One of the biggest reasons for big turnover there was burnout. By taking some of the workload off, by not having to make them deal with this procedural work, 9-1-1 operators are freed up to do their jobs. And that is the purpose of this AI.

Joy Powers: So that gets to something that I think can be a little confusing. We're calling it AI. Is this generative AI? Is that what we would describe it as, or is this a little different?

Adam Rogan: So this is different from, for example, asking an AI online, whether it's Chat GPT or something, to like, "Hey, make a picture of a purple duck flying in space." What this program is doing is: it's being trained on local information, whether there's parking laws in La Crosse County or court dates in Waukesha County, and able to respond to that using colloquial English language. And something I really foresee is we're going to be losing a lot of call centers and also a lot desk workers whose jobs are to work with the public or work with courthouse staff who are also answering questions in public. They're just gonna have an AI do it. We're gonna have a lot fewer of, when you call to get the pharmacy, you're not gonna have to listen to, press one for whatever and press two for whatever. You're gonna get an AI and you can just talk to it. And it will probably be just as susceptible to error as AIs are now, but it's not, let's say, wasting somebody's time who has a more important job to do. And that's where I kind of see the future of this type of technology going.

Joy Powers: I think all of us have dealt with AI at this point in some aspects of our lives, especially phone calls. And I don't find it very efficient, having had, I think, the experience you're talking about with prescriptions. AI at this point, when they're utilizing it on phone calls, is often pretty useless. It won't give you the right information. It will continually redirect you to different things that don't really make sense. And that's pretty concerning when we're talking about really essential services. Is that the future that we're looking at, or what?

Adam Rogan: So the public safety professionals who I spoke with through this story were very adamant that, when it comes to the emergencies, when it come to whether you're just calling in a car crash that you witnessed, or you're the victim of a crime, you're gonna continue to deal with a human. For the really important things, you're going to keep working with humans, and they said they can't see a future where human beings are taken out of that equation. So on a emergency basis right now, you're still dealing with humans. When it comes to these little procedural annoyances that we are continuing to deal with more and more in our lives, we're going to start seeing more AI. I think that's unquestionable. And I just find it very surprising and also interesting that it is kind of getting its first foothold, at least here in Wisconsin, through this public safety, even if it is not actually dealing with public safety situations, but rather kind of protecting public safety resources by answering questions that otherwise would have gone to courthouses or city halls who, oftentimes, the one person who has that phone is taking a lunch break, or if it's after 5 p.m. and nobody's there. Unlike a 9-1-1 call center, where it's 24-7.

Joy Powers: Have you spoken with people who have been using these systems?

Adam Rogan: Yeah, the initial views, the reporters who have talked to it have had, let's say, good experiences. But also when you're testing it, you tend to test them on predictable kind of questions. With these more complicated questions, I can foresee situations where you're talking in a circle with an AI. But at the same time, those more complicated questions are going to take up more of a staff member's time who is supposed to be answering 9-1-1. So how much time wasted? I can't say one or the other. It's a very specific use case.

Joy Powers: So why not hire a person to help with these calls who wouldn't be on an emergency line?

Adam Rogan: That is another question I asked, and one of the big answers was, you're getting so much turnover on these positions already. It's not the most desirable position, such that a lot of counties say, "I'd rather just pay the equivalent of that for this one program." And then also, that person doesn't need a vacation. That person's available 24-7. Aurelian is costing counties around $70,000, $100,000 a year. That's about one to two staff members, but also one or two staff members does not cover you 24-7. This kind of program does.

Joy Powers: But one to two staff members also have a lot more flexibility in how they might respond. As a person who has dealt with these kinds of phone systems, it's much easier generally to talk to a person and get to the department you're looking for, versus going in circles with an AI, potentially getting frustrated, not being able to get your issue solved, and then being really disenchanted with the entirety of how our government and community systems work.

Adam Rogan: I couldn't agree more. That kind of situation is going to continue to happen. But also, I think it's just, whether it's good or bad, the way that AI has ballooned, the way that the business systems and government systems in our country are going, I think we're going to start seeing more of this and not less. And hopefully, when things work well and efficiently, they get better, they don't suck up all the water in the world, and we can make a world more efficient, but for right now I think that we're very much in a trial period.

Joy Powers: So we'll see how this impacts government efficiency, I suppose.

Adam Rogan: That we will.

Joy Powers: All right, well, Adam, thank you so much for joining us here on Lake Effect.

Adam Rogan: I appreciate it, Joy.

Joy Powers: Adam Rogan is a journalist who wrote a piece on AI pilot programs in Wisconsin for Milwaukee Magazine. If you need some flexibility for when you listen, you can listen to Lake Effect on demand. Find it as a podcast wherever you like to listen, to download and enjoy on the go. Imagine your community being overrun by millions of frogs. It happened in the 1950s right here in Wisconsin. We'll tell you all about it next on Lake Effect, on 89.7 WUWM, Milwaukee's NPR. This is Lake Effect on 89.7 WUWM. I'm Joy Powers. It's July, 1952 in Oconto, Wisconsin. You wake up, stretch, pour some coffee, all while strange sounds filter in from outside.

Tea Krulos: It sounds like a finger being rubbed on an inflated balloon, so if you can imagine that sort of croaky sound, the air is filled with it, and also, sharp popping noises at times.

Joy Powers: Some witnesses compare the sharp popping to rifle fire, only it isn't distant rifles making these sounds. It's 175 million frogs. Tea Krulos is a contributing writer for Milwaukee Magazine who wrote about the great Oconto frog invasion for this month's issue. He sat down with Lake Effect's Sam Woods to discuss the phenomenon and how little of it is remembered today.

Sam Woods: Can you paint a picture. Like, what was reported from this time, and what did residents describe of what does 175 million frogs look like?

Tea Krulos: Yeah, so the story got some local attention in papers. But there was also a big magazine piece in The New Yorker, of all places, about the story in 1953. And they describe how, in the summer of 1952, there were millions of frogs pouring into Oconto near Green Bay, and the town was completely overwhelmed. I mean, the frogs far outnumbered the people. And so you had scenes like someone hearing chirping and flashing their flashlight into the backyard and seeing millions of glowing eyes kind of reflected back. It was kind of a grisly scene as people were running over them with their cars and mowing their lawns. You can imagine that was kind a mess. And then, eventually, it subsided and things went back to normal. But it was this kind of freak frog invasion.

Sam Woods: There's some skepticism on the number of frogs, so whether it's 175 million or just millions. But say there was indeed this big of a population boom of leopard frogs, what factors could lead to such a boom out of nowhere?

Tea Krulos: Yeah, so a great source for the story was, I was able to speak to Joshua Kapfer. He's from UW Whitewater, and he's co-author of a book, "Reptiles and Amphibians of Wisconsin." So I figured this would be the guy to talk to. And yeah, he did express some skepticism over that number, simply because we don't have a lot of good data from that era. He kind of pointed out that game fish and birds have been pretty well tracked. Reptiles and amphibians is not seen as a priority. So he said that this could have been possible at that time to have that large of a number. And the reason for it is he described frogs as being a boom or bust type of animal. So that means they lay a massive amount of eggs, because the environmental factors and predators will get a lot of those eggs. So they overproduce, so they can have a population. And usually, what happened in that area was the water level would drop and then it would rise, so a lot of the eggs would get wiped out. But this particular summer, that didn't happen. So all of the eggs were in pretty good shape, which is why you had this massive amount of frogs who emerged that particular year.

Sam Woods: The type of frog we're talking about is a leopard frog. What do we know about this frog and its relevance to Oconto's economy and culture beyond this fateful summer where we were just overrun with them?

Tea Krulos: Well, there's a lot of jokes because it happened at that time, frog legs were also very popular at bars. That was kind of a trend of the time for sure. Um, people were enjoying frog legs and restaurants and in bars, so they had plenty of supply that year.

Sam Woods: You mentioned in the article and The New Yorker talked about as well in 1952 that it was kind of, it's not like that this boom spurred a new industry in frog legs because it's, as you mentioned, it is kind of a boom or bust. It's not necessarily, you know, you get $175 million or so in one year, and then the next year is kind a normal or a bust year. And so, to set up your livelihood around this kind of thing isn't necessarily the easiest or most prudent business decision.

Tea Krulos: Right, yeah, and I did find an article from 1953. It was really short. It was just a couple sentences, but I think it had a picture of a frog and it said that the residents were relieved that the summer of '52 hadn't repeated itself, and they were overwhelmed by the frogs. But that must have been a disappointment to people who were maybe thinking it would be a trend and they could cash in on frog legs.

Sam Woods: You mentioned finding a short article about this, as well as the kind of larger New Yorker piece that it seems like kind of made this story famous or made the story live on. Tell me more about your reporting process of digging up this plague of frogs from over 70 years ago now.

Tea Krulos: Yeah. So I mean, the New Yorker article is the first thing you'll probably find if you're trying to dig into the story. And then I went to, you know, it's a good resource, newspapers.com, and just kind of did some keyword searches. And, you know, just found little scraps of stuff here and there. There's one really short article that it shows an Oconto resident, and it's hard to see in the photo because it's a reproduction or it's been scanned, but he's pointing at the window well of his basement, and it's full of frogs. So he's pointing that out to a reporter. So there's this like little tidbits of information, but not really like a cohesive story, which I kind of like, because it gives it a little bit of a air of mystery to it.

Sam Woods: I get that, and on the one hand it seems like 175 million frogs, this should have been in papers all over Wisconsin, if not the country. Like that is a wild scene. Did you encounter any skepticism, either at the time or afterwards, that this event happened at all given that you'd think it would be all over, you think everyone would be talking about it?

Tea Krulos: Yeah, you would. I don't know. I mean, things just kind of... News traveled in a different way back then. I didn't see any skepticism at the time. There were local Oconto area newspapers that reported on it. They just didn't make a big deal out of it, which I agree is kind of strange. I think the skepticism is about that 175 million number in particular, because I looked at The New Yorker article again, and I'll give you the single quote that references that. The author, Eli Waldron, says, "Frogs are difficult to count, but scientists and county officials in Oconto made a stab at it and estimated there was 175 million of them in the area." And that's the beginning and end of the information. So it's like, who were these scientists and county officials? Like, how did they come to that number? There's not really any information on that. So, again, like, what I've been led to believe is that it certainly would be possible at that time. It wouldn't be today because, sadly, leopard frog habitats have been in sharp decline since the '60s at least. And that has a lot to do with development, predators, including invasive species. So there's no way that would happen these days. But in 1952, it certainly was possible.

Sam Woods: How does this story live on? How does Oconto or Northern Wisconsin remember this?

Tea Krulos: It's a story that seems to kind of get dusted off perennially. Like, every once in a while I saw like Discover or Discovery Magazine had done a story on it. It gets dug up once in awhile and someone will do, you know, a piece on a website or something about it. There's no, like, thing to commemorate there. I mean, I always love it when a small town will have a festival celebrating their local lore, you know. That's what makes a place unique, but there's no frog invasion festival. There's no frog statue, unfortunately. There is a country lane called Frog Pound Road, and I was told that that was inspired by this incident, but that's about it there so far.

Sam Woods: I can imagine how the story comes back perennially, given my reaction when I saw the description of the story that you had written, not even the story itself. And I was immediately captivated as like, holy moly, like what, what are we even...am I reading this right? There's no way.

Tea Krulos: It's a very strange little chapter of history, for sure.

Sam Woods: So I want to ask you, Tea, before I let you go, speaking for yourself, do you view frogs the same after this story?

Tea Krulos: I never really wrapped my head around seeing so many of them at once. I can't imagine what that was like. I mean, just imagine you walk out your front door, and the entire surface area of the ground is just completely covered in frogs hopping around and croaking. Like I don't know what I would do in that situation. You know, I would be afraid that... I wouldn't be afraid, but I wouldn't want to leave the house, just because, I don't know, I don't want to step on a bunch of frogs everywhere I go. So it must have been a really weird, surreal time in Oconto.

Sam Woods: Well, until the next frog invasion, Tea Krulos, thank you so much for joining me on Lake Effect.

Tea Krulos: Thank you.

Joy Powers: Tea Krulos is an author and contributing writer for Milwaukee Magazine, who wrote about the great Oconto Frog invasion. He spoke with Lake Effect's Sam Woods. And that wraps up today's show. I'm Joy Powers. If you ever miss Lake Effect, you can find all of our conversations at wuwm.com. If you'd like to take the show on the go, you can download the Lake Effect podcast wherever you get your podcasts. Tomorrow on Lake Effect. We'll learn about the underwater archeology program at UW-Milwaukee, and what it explores and teaches students. Thanks so much for joining us right here on listener-supported 89.7 WUWM, Milwaukee's NPR.