Rich Lavers at Google Cloud Next '24

Speaker A: Welcome to our special edition of Scoop News Group's AI Week podcasts. I'm Wyatt Cash and we're here at Google Cloud next in Las Vegas. Uh, and joining us here in the public sector booth on the expo floor is Rich Lavers.

Speaker B: I'm Deputy Commissioner for the New Hampshire Department of Employment Security. Just to give you a quick sense of what my agency is responsible for. So we manage the state's unemployment program. Uh, we also help individuals find employment. So working closely with, with employers and job seekers through big events like job fairs, small events like one on one, um, interactions in our uh, One stop centers throughout the state of New Hampshire. And then we also manage the state's labor market information program. So um, looking at unemployment rate, labor force, occupational projections. So a lot of different programs fall under our umbrella. Um, and our focus here has been really on the unemployment program and how we can um, improve that with lessons learned coming out of the pandemic.

Speaker A: Talk to us a little about um, how AI has impacted your efforts at where you work, uh, and kind of where you're putting AI to work, uh, for the state uh, of New Hampshire.

Speaker B: Yeah. So I think the most exciting application uh, for AI, at least in our world, um, thus far has been in the way that we make decisions on eligibility for unemployment benefits. Um, you know, whenever you're dealing with an economic downturn, um, you're going to have situations where you're not going to have enough people, you're not going to be able to get information fast enough. And I have never met a person who has received their unemployment benefits too quickly. Right. So you're always going to want those highly trained, skilled individuals making the right decision decisions. And um, what we've found with AI is that we can improve the way that we're um, um, obtaining the information from the customer and we can improve the way the information is presented to the worker who's making that decision on whether someone's eligible or not for those benefits.

Speaker A: And uh, we're midway through uh, Google Cloud next here, uh, talk to us a little bit about any of the announcements or uh, other um, sessions that you've attended that uh, where you kind of came away with some ideas that you plan to take back with you.

Speaker B: So um, what I'm, I'm really impressed with the focus uh, that Google has been putting on public sector. Um, you know, I say that from, you know, I'm a 14 year public sector guy. So um, that's where my passion is. Uh, so I love to see that investment in public sector But I think the investment in public sector by organizations like Google is incredibly important for the future growth of life changing products like AI. Um, because I think a lot of people still have some level of skepticism for AI, maybe a little bit of a trust issue. And particularly the generation of people that tend to be in positions of authority and state elected bodies and federal elected bodies, they're not used to this type of technology when they were, you know, when they were growing up. When I was growing up, you know, the phone was connected to the wall, right? And so that technology is not as, um, just you know, saturated throughout your life like it is for younger generations. So I think through looking at government programs, particularly government benefit programs like the unemployment program, bringing in AI to improve that program helps to increase people's exposure for what the possibilities of AI are. And I think that will help with the trust factor that will reduce skepticism. And it's only going to help down the road as those people who are looking at regulating this new body of technology and doing regulations that make sense and aren't too restrictive on the technology. So I think it can do a lot through enhancing and improving government programs and how we deliver services. I think can do a lot to grow AI, uh, for future use.

Speaker A: Um, one of the things we heard here was that um, regulations on the books that tended to dictate how public servants need to work with the public can be constraining when it comes to applying tools like AI. Um, can you talk a little bit about ah, are the regulations holding you back and what needs to happen to make sure you can innovate the way you want?

Speaker B: So um, you know there's a whole set of um, regulations that come into play. Right. And New Hampshire is one of many states, I'm sure, and I haven't surveyed the whole country, but there are legislative proposals right now, uh, to put in controls around how public agencies are able to use AI. Um, some of it makes a lot of sense and what I've seen, you know, making sure that the customer knows when they're engaging with AI, I think, I think that makes a lot of sense. I think people appreciate that awareness. Um, and I think those types of requirements at the state level or the federal level will help to grow AI, will help people to have more confidence in dealing with organizations that are utilizing AI. But if you look at other types of restrictions on areas where maybe AI is going to be prohibited from being utilized, I think that can be dangerous. And um, I hope that you know, more um, uh, states will stay away from getting into that area and more, uh, focus on how we can, you know, accept the fact that AI is here. How can we better leverage it to improve the way we deliver services to the public? Because let's remember, you know, for the services that government delivers, they're inherently not profitable. Right. Or else government wouldn't be doing them. But no one else wants to deliver the unemployment program, um, so it's left to government. And we can't just simply continue to have people lower their expectations to come into an antiquated system just because it's run by government. I think we can do a lot better to deliver a more modernized product for, for individuals and it can improve their experience and you know, can show that government, um, can leverage AI just like private industry can leverage AI. And it's going to become, it's going to help introduce more of the public to the power of AI and the power to do good.

Speaker A: Great point. And as someone said at one of the sessions this morning, I think there's a lot of opportunity for government actually to lead the way in this area to improve the kind of efficiencies that the public naturally expects but doesn't typically expect of its government. Right now. Talk to me about what's next on your horizon. What are you hoping AI will help you accomplish in the next couple of years?

Speaker B: Um, so New Hampshire's got a great opportunity right now. We were selected by the federal, ah, Department of Labor for a fairly large modernization grant. So, um, to take our existing unemployment system and learn from the pandemic for the areas that we can make not only the customers experience better, but all the experience of our staff better as well. Um, because if you think a lot of the focus for what AI can do for government programs like the unemployment program is really focused on how can we improve that customer experience. Right. But there's also another side to that and how can we improve the experience of the workers who are charged with, um, obtaining the information and then making the right decisions in adjudicating those benefit claims. And if we're not able to modernize that process, we're going to continue for government as an employer. We're going to continue to have a major retention problem for our own workers if we're not giving them the opportunity to work with the latest and greatest products. So I look at what we're doing with AI, um, with our AI Powered Adjudication Assistant with um, Google Public Sector as a great learning opportunity for other states, not only in the workforce space, but I think, um, if you're in the health and human services world, if you're in the higher ed world, I think there's a lot that you can look at from this and see an application with the way that you're currently obtaining information and how your staff are processing that information. Just a simple example of that. We've forever, we've decided ahead of time how we're going to ask you the questions about your job separation. We don't need to do that anymore with LLM. We can just ask you to tell us about it. Right. We can pluck the information or use AI to pluck the information that's relevant for presenting that to a human being to evaluate the data. And we can present it to them in a much more succinct way. Um, they can do the due diligence to make sure that AI is pulling the information correctly. But I don't need to have the questions preordained anymore that I'm going to ask you about your separation. I'm just going to have you tell me about it and let you tell me the story. I'll let AI do the work of pulling out the information that's relevant and what's not relevant. And we're going to get to a place where it's a more accurate decision in a much more timely fashion than what we're currently doing.

Speaker A: That's a great example. And I think one of the points you made that I found interesting is I think attracting, uh, the next generation into public service. They're not going to be wanting to look at green screens. Uh, they're going to expect this kind of technology, I'm sure. So the sooner, uh, state and local agencies can get it implemented, you know, the better chances you have of, you know, uh, attracting this next generation of workers. Which leads me to my last question. You know, I think the technology is taking off like a rocket. I, I don't know how anybody can keep up with all the new developments just announced here. What are you doing to uh, up the learning curve of your staff? Uh, so they're able to really understand the use of AI and take advantage of it.

Speaker B: Yeah. So we've been fortunate with the piloting that we've done with Google thus far on our adjudication assistant. We brought in a lot of our staff, uh, to be on this from the ground up, um, so to help them develop the product. And I think going forward, because you know, we're looking at an AI powered adjudication product that's going to continue to learn, it's only going to Continue to learn through the hard work of our staff. Right. So making sure that the questions that are being asked of people on that when they come in to file that initial claim, sometimes AI is not going to ask the right question and sometimes they are and helping it learn which ones are good and which ones are not, I think is going to be a role of our staff and it's going to really help like we just talked about, I think not only on the retention side, but as that gets out, that's going to help with recruitment. Right. And giving people the opportunity to work with a product that they can be proud of. That when they're, you know, at that party on the weekend and someone asks what you do, they can talk about, you know, this great new innovative AI powered process that they're uh, a part of. And they're probably going to see a lot of kind of eyeballs get really big where people are going to be receiving that, particularly non government people and be like, wait a minute, I thought you worked for government. You know, why are you doing something innovative? Right. I think, I think people can be more excited about that. It will really help us retain, um, what we know is a, a large group, a talented group of younger generation. They want to go into public service. We need to give them, um, kind of the reasons to do it.

Speaker A: I love that vision of the future. Well, Rich Laver, thank you so much for joining us here at the Google Cloud Next Summit here in Las Vegas. Really appreciated you stopping by and filling us in a little on what's going on in New Hampshire.

Speaker B: All right. A great opportunity to chat. Really appreciate it, thank you.