2025 State of Gov Tech - Innovation Launchpad Winners

So hey everybody, uh, my name is Steve Townes. I'm VP of Content Strategy for Government Technology and eRepublic. Uh, my pleasure to host a special edition of In Case You Missed It today. So, uh, over the past several years, Government Technology has partnered with CivStart to sponsor the annual State of GovTech event, which showcases potential innovation in the public sector It inspires the audience to think creatively about using technology to address pressing issues and deliver high-quality services to their communities. Each year, promising GovTech startup companies are invited to pitch their companies and their solutions in front of a panel of judges, which includes government employees, industry experts, and GovTech investors. So 24 separate companies were evaluated by the panel this year. And 3 startups were chosen as winners. So on today's episode, we're giving each of these State of GovTech winners, that's Magic Notes by Beam, Chief AI, and Madison AI, an opportunity. They'll all get an opportunity to introduce themselves and present their GovTech solutions today. So welcome to the show.

Thank you.

And let me first turn to Alex Stephanie, who is founder and CEO of Magic Notes powered by Beam. Alex, welcome. Glad to have you here.

Great to be here. Thank you for having me, Steve. Fantastic. Well, let's kick off. We're going to talk for about 5 or so minutes about Magic Notes, which is really about less admin and more connection. But I'm going to just start by offering you an idea, which is that to have any chance of changing the world, you need to be able to change at least one person's life. And for us, that person was Tony. And Tony had been unemployed for 21 years. He had been homeless for large chunks of that. He'd been in and out of prison. He'd had all kinds of different addictions, and really he was at his lowest ebb. And I met him when I was starting a technology-powered human services organization. And I sat down with him and he told me that his dream was to become an electrician. And we supported him to get his electrician training. And for the first time in 21 years, he was in work. And not only was he in work with all of the structure and financial freedom that gave him, but it actually allowed him to be reconnected with his family. And over the coming years, we pioneered smartphone-first social services in the UK and helped a lot of people. In fact, in the next 5 years, we supported over 6,000 people. Into jobs, homes, and, and we gave them new skills through training programs. However, there was a big problem. By 2023, our frontline team was burning out with admin. They had manually written and typed over 600,000 case notes. That was huge amounts of time that was taking them away from actually helping people And it was also sapping their morale. And so really, we thought there had to be a better way. And fortunately, there is, and it is called Magic Notes. Magic Notes is a tool that creates exceptional compliant documents, and not only any documents, the highest quality and most consistent documents. And it's very, very easy. It's really as easy as 1, 2, 3. Let's, let's walk you through quickly how it works. Someone is having a conversation in a frontline service and they just click record when they start the meeting. It's a big pink button and off they go. They click again when they stop their meeting and then they pick the document type and within a few seconds they have a very high-quality document. And we have built more than 1,200 different social services documents for more than the 120 government partners that we serve. Now, over 60% of UK social services teams use Magic Notes. It really has become the tool of choice because it's been able to reduce admin time so profoundly. This is actually a stat that was— that comes from independent research from a well-known academic in the field of AI, Professor Rob Proctor. He's from the Alan Turing Institute and the University of Oxford. Professor Proctor found through his study that support workers benefit from better client engagement, time savings of 50% or more, and improved quality. And that's a really important note here. which is that these tools are not only just about saving a day, sometimes 2 days a week on admin, it's also about allowing more human-centered services where people can actually look someone in the eyes and really connect with them as part of a social service. And of course, being so widely used, the tool has made a bit of a stir in the UK. We've had meetings with the Prime Minister and a lot of major politicians. And major publications like the BBC have called it a game changer. But above all, it's loved in the field by thousands of caseworkers. And here are just a few of literally 500, 700 quotes you probably have by now of people raving about this technology. I'm so incredibly happy with Magic Notes. It saved me so much time. The best thing for caseworkers ever. It's empowered me to excel at my role and has made a profound difference in my life. And overall, through thousands of ratings, it's been rated 4.5 out of 5. Where we are today is we are onboarding our first US government customers who are mostly but not exclusively counties. We have several slots left to onboard pilot customers in 2025. If any of you work in human services or you have colleagues or friends who work in human services, we would love to show you what this technology can do. This has been a very, very brief overview. It can also make complex reports, provide insights, provide analytics, and do a whole bunch more. So, uh, do feel free to reach out to me. My email is very simple. It's just Thanks very much for listening.

Hey, Alex, thanks a lot. That's great stuff. Thanks for telling us a little bit more about it. So next, let me bring up Shannon Beckham, who is founder and CEO of Chief AI. Shannon, over to you.

Great, thank you. Hi everyone, great to meet you. I'm Shannon. the founder of Chief AI. We build software to power the world's leaders, and today I'm going to show you what we're building. For those of you who have worked in a government office, you know that it can feel something like this. Staff rely on paper to manage just about everything, from the schedule to briefing materials to sending and reviewing things through your inbox. Here's an example of what I mean. This is what it takes to schedule and prepare for a meeting, both for a mayor And also the President of the United States. The team receives the request via email. They then gather more information in a Word document. The staff then type that out and send it around via email to get input. The schedule gets sent. A memo is handwritten, reviewed by multiple people, approved, and then gets printed and put into a briefing book that gets delivered each night. And then often the schedule is constantly changing, so updates get sent around. And finally, the principal can review the memo and walk into the meeting. This process happens hundreds of times each week, and I know this because I've lived it. That's me right there staffing a US senator. I've worked for leaders across the government and private sector, and in each case, this was the manual process for all of them. Every time a step in that process breaks, an opportunity gets missed, A relationship can suffer, or a wrong decision can get made on outdated data. And ultimately, this problem is compounding over time because none of this institutional knowledge is being stored in a single place. So I really set out to solve this problem. How can we help all of the government, government leaders and their teams across the country be far more efficient and strategic? And when I was looking in the market, there was no software product built to manage executive leaders, especially in government. And that's why we're building Chief AI. Chief is a command center designed to manage a leader's time, relationships, and priorities. We start with time because ultimately your schedule is a reflection of your priorities. So we completely reimagine with leaders how they're building their schedule. We then store all of that data in a relationship database, and we use that data to help give you AI-powered insights to move toward your goals. I'm going to walk you through an example. This is how we've been working with the mayor of Denver to take a process that once looked like all of that messy paper and that 12-step scheduling process to being one that's now powered by Chief. So we start with scheduling. We created a dashboard where every scheduling request flows into a single place, from online forms to emails to text messages. And for the mayor of Denver, when we started working with him, His team had a backlog of hundreds of requests sitting in a Salesforce backend. Immediately, Chief went through, triaged those requests, and summarized and categorized all of them so that the team could see whether they've been— whether he's met with that person before, whether aligned with his priorities, and how to make it on the schedule. We then use AI to help generate the memos for each of those meetings and events, and all of those memos get compiled automatically in a daily briefing book. That has the schedule, his tasks, and all of those briefing materials. That data is then saved in a relationship database, so you can look up any person or organization and see the last time you met with them, their kids' names, what policies they care about, all the information you need in order to walk into a meeting prepared to really drive toward your outcomes. There's also a dashboard that provides real-time insights on his— how he's spending his time against topics, certain geographies, and just from meetings to events, really understanding how he's spending his day. Most importantly, his team can also now ask Chief a question from anywhere on the platform. And so for the mayor of Denver, this is questions like, I'm hosting a roundtable on economic development. Who are key stakeholders in my community that I should invite? Chief not only makes you more efficient, but it's also making him more strategic. And most importantly, this is all purpose-built from government for for government from day one, so that your data is secure and compliant. As Mary from the Denver team said, Chief AI is game-changing. It removes hours of manual process that I've done for years. Every principal's team needs this product. And the truth is, across every single one of these scenes of mayors, city managers, governors, and any government leader. Behind the scenes is a merry. There is a team that's working hard to do admin, respond to constituents, and set up each of these events to help them drive impact in their communities. We know that most tech companies tell you they're building software to change the world, but at Chief AI, our mission is to build it for the people who actually do. And that's what we're doing here. We launched just earlier this year, and we're bringing on our next set of pilot customers this fall. If you are a government leader or a team member who wants to pilot our program, we would love to work with you. You can reach us through our website at getchief.ai. Thank you.

Thanks, Shannon. What a great solution. Appreciate hearing a little bit more about it. Now let me call up Erica Olsen. She's CEO of Madison AI. Erica, welcome.

Awesome. Steve, thanks so much for having me and having all of us. It's a pleasure to be here, and thanks to everybody who's tuning in. What are we up to at Madison AI? Well, Madison AI is an AI engine, as we like to say, for everyday work in cities, counties, and water authorities. We named ourselves after James Madison, who believed in good governance, and we believe in good governance because Good governance, great governance makes great government, and great government makes great communities. And that's, you know, that's what we're really all about. We're specifically focused on internal staff work and admin work and knowledge work with a specific and express purpose of saving staff 5 hours a week every day on repetitive tasks. And that's really what generative AI can do for all of us when applied correctly and purposefully. And just like my other colleagues that presented, Madison AI is purpose-built for local governments specifically to deliver time savings and quite frankly be a mission multiplier for all cities and counties across our country and hopefully across the world. So let's jump into what we're doing differently. So a month in the life of local government looks like the next slide. So Looks like this. Some broad areas of functions within local government. There's many more, of course, but within management and clerks, planning, budget, and also with the elected officials and human resources, there are like hundreds, if not thousands, of consistent, repetitive, important tasks that staff within these functions have to do, must do, and need to do to run a great government. And this is what it looks like when you add it all up. This is based on a city with 250,000 citizens, so you could scale this up or scale it down. But fundamentally, it's a lot of time. It's important time, but it's also administrative time. So things like staff reports and board memos, or for planners and engineers, things like zoning verification, or for budget and procurement, we're gonna look at an example of this, completing and submitting a procurement request, which is so important and, and something that that everybody loves to hate. For elected officials, our elected officials are prepping for their meetings either each week or every other week. And those are huge agenda packets that are put together and need to be read. And then of course, human resources is always fielding tons and tons of questions and responses that are important but could be assisted through AI. So with Madison AI, what we've done is we have taken all of the institutional knowledge of the agency and contained it in a closed, data lake. And we're essentially like a custom GPT, but not with the World Wide Web, with just your information so that you can and folks and staff members within each of these functions complete these important, but quite frankly, time-intensive administrative tasks in 20% or 80% less time or saving, saving 80% of what it would really take to get, to get the task done. So what does that really look like? Let's check it out. We have a video to show you. In the City of Corona, we are currently piloting what is a procurement request AI assistant. So it's— this video is going to run for a second, and you can kind of watch it as it's doing its thing. But behind the scenes of this AI assistant is just the City of Corona's purchasing policy. And built into this is the 5 different procurement requests that are required based on whatever it is you have to be procuring. This is an example. And it's going kind of fast because, of course, it's an AI assistant. We know how chat works. But what this is doing is it is feeding the requester, in this case, a series of questions to walk you through and help figure out what path you have to go down as a requester to procure these computers, which the first thing is that everybody at Corona is super excited about is there is an objective path for which procurement request is required that we're not like asking you having to do a bunch of rework and debate which, which request, which happens all the time. And then at the end, once the AI assistant has guided the requester through a series of questions to help actually fill out, you're seeing on the screen now the actual RFP that is filled out for procuring these computers based on the specific template that is required to comply with the purchasing policy of the City of Corona. So what's important about all of this is it's just the City of Corona. It's just their information. It's very much to their standard and to their requirements. And then also, of course, it's completing the paperwork, which is awesome. When we move this into full production, of course, you'll output it, and that would be submitted into the, into the purchasing request system. So we can jump to the next slide. So kind of knowing that, that's just an example of some of the things that we're able to do when we're building with Madison AI. So what does it really look like? So here's the estimated time savings for the City of Corona in a month. They do about 9 procurement requests. That was at least last month, the large-scale ones, not the small-scale ones. And add it all up, very conservatively sitting around 350 hours a month. And I just want to point out about this, you know, the dollar saved is really more about shifting time.

Mm-hmm.

You heard that from Alex too, which is so cool. Shifting time from doing paperwork and administrative work to doing people work. So to really actually taking the time to build a purchasing request well and to scope it well, and quite frankly, to reduce the amount of rework and frustration and negative job satisfaction of doing such an action. So that's really what Madison AI is all about and what we're really enabling with, with an AI assistant. So the next slide. Just shows you a little bit of a diagram of how it really works. And just to summarize, all of the information on the left-hand side is just the City of Coronas. We move it all together and we do purpose-built AI-assisted workflows that then output on the right-hand side very specific outputs to that particular jurisdiction's templates. So in conclusion, just the last slide. We are really focused on quietly remaking how work gets done within cities and counties. And you can see we're in about 30 cities and counties at the moment, about 50 pilots in the marketplace. We too, of course, are onboarding folks this fall. So if you would like to make work better in your organization and reduce administrative tasks across the functions of City, the Office of the City Manager, electeds, clerks, planning and zoning, budget and procurement, give us a shout. We'd love to talk to you. Thank you.

Erica, thank you. Really appreciate the look at what you guys are doing as well. And, you know, thanks to all 3 of you guys. I mean, that was an amazing look at the innovation that's going on in this space and really appreciate what you're all doing to support states and localities that, as we all know, have a really tough and seemingly tougher all the time job, right? So great examples. Maybe as we close out today, I wanted to let each of you guys kind of leave one last thought with our audience. I mean, really, what's one thing you want this market to know about your company and your solution? I'll give you each a chance to answer that. Erica, I'll go to you first and then we'll run around the screen here. But Erica, give us one bottom line takeaway.

I think a bottom line takeaway for us is how do we fit in the landscape of AI tools that everybody's evaluating today? And what's important in the domain that we're occupying is we're really focused on, like, staff work and really empowering staff. And the other piece of that is purpose-built for local governments. We are not general purpose. Copilot's general purpose. Madison AI is purpose-built vertical AI specifically tuned and built for the jurisdictions and the functions that I walked through. And that's really the difference maker in making AI actually deliver on the value proposition position of being a legitimate assistant to get work done.

Thanks, Erica. Shannon, same question to you. What should people walk away knowing about Chief AI?

Great. Building off what Erica said, for Chief AI, our differentiator is that we are working with executive leaders across government. So if you're an executive leader and you have a chief of staff, a scheduling team, others supporting your office, we want to work with you. And I think the other thing I'll say is just a shout out for working with AI startups. You know, we're in this era of rapid innovation, and I've sat in your shoes in government offices, and there wasn't always a clear path on how to work with new tools. But right now we're seeing a complete shift, and I've been so amazed by all the government employees and teams who are willing to pilot and partner new tools. And that's what Erica, Alex, and I are all here to do, is work hand in hand with you to really purpose-build these tools for your work in a way that's safe and not just retrofitting something else that's coming from the private sector. So I would just give a shout out to everyone here. And I think that that's this moment that we're in that's really different than one we've ever seen before.

Yeah.

Thanks, Shannon. Alex, you get the last word. I'll try and be—

I can't believe people are still doing that after God knows how many years of COVID But anyway, I think the thing that I would love people to know about Magic Notes is that it's the only tool for social services teams that is actually built by a social services organization. We deliver these services ourselves, and as I mentioned earlier, we built it originally for our own team, and we refined it and refined it and refined it, and only then did we get it out to the market where today it is used by almost 2/3 of social services teams in the UK. And it's because of that success in the UK that we've opened our New York office and are so excited to start working with our first US partners out here. But what does that really mean in terms of the tailoring? So the first is that social services meetings are often happening in real life. So Yes, it integrates with Teams and Meets and all of these things, but it's primarily built for in-real-life meetings, which means it can handle any handset, any browser. It can handle chaotic background noise, kids screaming, multiple languages. That's where we've really fine-tuned that tech, and so it really excels in that environment, unlike generic tools. It also means 1,200 social services templates. social services analytics, and really importantly, on-the-ground adoption and training. So, you know, one of the things that I'm sure Erika and Shannon could testify for is it's not enough to have amazing technology. You also need to support partners to successfully adopt that technology. And that's also a really important part of how we work. And that's why we now have a US footprint. And I'm personally a CEO relocating to, to the US. So Yeah, that's it from us, the only tool for social services teams built by a social services organization.

Well, thank you, Alex, and thanks to all of you guys, and congratulations once again. And thanks for sharing, you know, a little bit about your vision for GovTech kind of going into the future. Like you said, it's an exciting time and there is a lot going on, so we appreciate hearing from you guys around what you're doing. We'll have links to all of these startups for more information in the episode write-up that'll appear on govtech.com. So we wish all you guys the best, and until next time, have a great weekend, everybody. Thank you.