Music . The first item of business is our approval of the April 28, 2026 committee summary. Are there any additions or corrections? Okay, seeing none, I'll take a motion to approve. Second. I heard a motion to approve from Council Member Civigni, a second from Council Member Curtis. Any discussion on the motion? Okay, all in favor say aye. Aye. Any opposed? Okay, that motion passes. For our presentations today, the first presentation we have on our agenda is the review of the artificial intelligence policy. Council Member Curtis, this is your item, so I'm going to turn it over to you for an introduction. Thank you, Chair. And, yeah, so, colleagues, this item stems from some conversations that I had during last year's budgeting process while serving on the general government and services budget link. I was largely in agreement with my colleagues about the budgeted items we were approving, but there is one item that I struggled with and continued to struggle with, candidly, involving Microsoft Co-Pilot, which is a so-called artificial intelligence companion. It might seem innocuous to some folks, but it was a red flag for me, and I wanted to dig deeper into the issue. Like many Lexingtonians, I have profound concerns about AI, its devastating environmental impacts, the threat it poses to data security and the means by which it's been forced into everyday life by a handful of billionaires and corporations. At my core, I believe that AI has far greater potential for collective harm than it does for collective good. And I believe that this potential for collective harm increases exponentially when these technologies are utilized without intensive scrutiny and rigorous human oversight. I agree with Pope Leo when he wrote the following and his recently published encyclical Magnifica Humanitas, that it is not enough to invoke ethics in the abstract. Robust legal frameworks, independent oversight, informed users, and a political system that does not abdicate its responsibility are required. Otherwise, change will be governed only by technocratic thinking presented as necessary and inevitable, ultimately imposing rules shaped by those who control the data, infrastructure, and computing power. I sponsored this item, and I'm proud to champion this policy, not because it wholly aligns with my personal beliefs about AI, because, spoiler alert, if you can't tell, it doesn't, but because I know that these technologies exist, and it's no longer possible for us to avoid their existence, and they are rapidly evolving. We have a responsibility to our constituents to educate ourselves about them, educate the public about them, safeguard their usage by our government, and to be transparent about when, how, and why LFUCG is utilizing AI. We cannot abdicate that responsibility. I'm deeply appreciative of CIO Rogers for being here today to present to us, as well as for her and her team's thorough and thoughtful work on this policy. So with that, I'm going to turn the microphone over to CIO Rogers. Thank you, Council Member Curtis. And although my sphere of influence is probably much narrower than the Pope's, we'll try to do right by that ethos. So we'll start with a little bit of background here and trying to simplify an overview of everything this is and honestly everything it's not. It's pretty difficult, but cranked out this little infographic with the help of Microsoft Co-Pilot, full disclosure. So artificial intelligence as a field, as a scientific field of study, was established way back in the 1950s. A couple decades followed with experiments in robotics, simple algorithms for things like game play. And you'll remember Deep Blue beating Kasparov at chess in the late 90s. A lot of experimentation over the decades that sort of narrowed the scale a bit. Machine learning around the turn of the century brings big data, if anybody remembers that term, and the big hardware that came with it. So now the learning becomes much more advanced because the data sets are bigger. They're more expansive. There's a lot more data at the tool's disposal readily available. That's more along the lines of Watson, IBM's Watson, winning Jeopardy over two former Jeopardy champions. It's getting much broader and deeper knowledge. Deep learning, the next little concentric circle down here, uses multilayered neural networks to process specific types of data. So images, speech, specific things for the purpose of training that model to recognize patterns in those things. So then around early 2020s, 2020 GPT-3 is released. JAT-GBT follows in November of 2022 and thus begins the Gen AI boom. generative AI uses those neural networks to perceive, to learn, and then to generate new content. So it takes all of that data, lets it churn a little bit, creates new data in form of content, text, images, and the like. So we are given at that time numerous large language models, obviously speech, text, Diffusion models for images, multimodal models for audio, everything under the sun now can be created with AI. And then it didn't take long for those to form into what we would know as agents. Agentic AI kind of takes form when you start to pair some of those things together or assign human tasks to some of these models such that they operate autonomously and adaptively. So, for the terms of this policy, these are the definitions that we're using. So, AI, artificial intelligence, sort of our inclusive term for these technologies. You can see, I think I paraphrased these a little bit from the policy document, which you do have, but it's essentially the whole thing. It's the concept, machine learning and related technologies that use data to train statistical models so that it can perform tasks. Generative AI then being an artificial intelligence system that's capable of producing content synthetically. So Gen AI creates things. The models recognize patterns and data structures and then use that data to generate new data and content. A genetic AI, our AI systems capable of autonomously accomplishing tasks without human involvement. So message send, code execution, code development, or other types of tasks all fall under that umbrella for the purpose of this policy. So agentic AI, essentially an agent of your business, essentially an employee. So why do we need a policy for this? Well, we have policies to manage our employees. We do have CIO policies to manage how technology is utilized across the government. So this kind of falls into both of those domains. Our foremost responsibility in the Department of IT is the stewardship of LFUCG's technology assets, our infrastructure, our data assets, particularly when that data relates to our residents or our employees. So we seek through these policies to manage risks. The risks in this particular arena are somewhat unique. We're not just looking at your typical cybersecurity risks. We're not just looking at data privacy matters. We're also looking at potential reputational harm. We're looking at bias that's built into these models. We're looking at risk to our public trust. So we wanted to make sure that we got out early with some policy directives to make sure that people were, you know, not only using the right technologies, which we can sometimes control, but also using them in the right manner, which we often are limited in our control of. So the scope of this policy is broad with that understanding. We're looking at data, and sometimes that is handled. The number of systems is innumerable, the systems that contain our data. We're looking at our business processes. We're looking at our employees. We're looking at all of our infrastructure, be it endpoint devices, networks, business applications. So the scope is terribly broad, and we really manage by exception, which is kind of deliberate. But the directives, in short, are all of these risks must be addressed. Technology is a means, not an end. Work product remains the responsibility of employees. They must check their work for accuracy, bias. You know, slop is the colloquial term. And also that acceptable use does not imply appropriate use. The use cases must be evaluated, not just, you know, this tool is approved, I can use it for everything. And we'll get more into that in this next slide. Acceptable uses include things like things that writing assistants do, drafting content support. We have a lot of documents here. We produce a lot of documents. We read a lot of documents. Sometimes it's nice to have a tool that can kind of summarize or help draft those for you. Process and workflow efficiency. We've been doing this for a long time. We've been automating processes for as long as we've had technology in this organization. So those are always good opportunities to really examine the process, see if we can streamline. Data analysis and research support, again, finding patterns and data that's accumulated over time, really good use case for these tools. And then employee training and enablement, anything that helps folks experience simulations, anything that helps them experience something that they will encounter in their jobs, definitely support that type of use case. However, there is no low risk use case with this technology. This is why we have these conditions in place. First, human accountability always forever. That's human in the loop is the term for that, but the output from these tools has to be reviewed, again, not just for accuracy, also for completeness, appropriateness, lack of bias. Make sure that it's compliant in terms of what you would want to produce for a third party or the public or a regulator. Also data protection has to be top of mind. No sensitive data is to be employed in tools without approval. That includes using it to prompt. So you can't feed anything sensitive or anything specific into these tools. So anything, again, that's intended for residents, regulators, any external audience really needs to be reviewed and disclosed pursuant to applicable policies. Applicable policies is another really kind of broad term here. That can include our CIO policies for the technology, CAO policies for employee behavior, but also codes of ethics for particular professions. Think in terms of attorney-client privilege, and then, of course, all the sensitive data types that we handle, personal identifiable information, personal health information. We've got information about children who participate in our programming, so always treat those as sensitive information. And then for disclosure, if you have to, for your use case, disclose that you have used AI, you've got to do that. So that's one where we really rely on some of our leaders outside of IT to advise their staff on when that's appropriate and what that needs to look like. Prohibited uses, everything here, plus whatever your boss says, don't use AI for that. So in general, we want to make sure people are using approved tools. Commercial AI products do not have data protection capabilities built in. So we really want to discourage people from using those things. Again, inputting or prompting with sensitive or customer-specific data, that is a huge data leakage risk. You could be held personally liable. Autonomous action, this, again, points to the agentic behaviors without direct human oversight. So, again, making any kind of decision that's going to impact a resident or performing a task on behalf of a staff member is something that has to be very closely monitored. And, you know, fact check and quality controlled over time as well. Any type of impersonation or misrepresentation, full stop, not good. You might recall that our government communications office has advised against using AI produced art in social media posts and newsletters and things like that. So again, that's acceptable use, producing imagery. It's not advised, it's not appropriate use of the technology based on other guidelines that we have in place. Again, production of any kind of unsafe or misleading output that could cause us any type of harm. Again, reputational harm and damaging the public trust is top of mind with this one. And then bypassing corporate controls. If you've got a VPN that makes it look like you're not part of this government when you actually are, and you use that so that you can use a commercially available AI tool, that's a major no-no. So again, this is not a conclusive list, but sort of the spirit of what we're trying to do here. So enforcement wise, we do have some technical controls available. When you sign your acceptable use policy, when you get your computer from the LFUCG, you are told that, you know, assume that all activity is monitored and logged. Because it is. Now, do we have time to look at all those things? No. But we do have the capability to query whether or not people have engaged with these tools in ways that they shouldn't. So we do have that monitored. We are logging those behaviors. We do plan to block some of those commercially available tools. I didn't really want to do that until we had an acceptable enterprise standard in place in their place. So now that we've had that, now that we've had a little bit of literacy training there, we'll probably be blocking some of those tools. Again, failure to comply, whether it's intentional or through negligence, whether you know about it or not, could be disciplinary action or we could take away your tool. So a couple of, you know, you misbehave with the tool, you don't get to have that anymore. Again, violations here that involve any sort of legal repercussions, any sort of impact to a third party that has to be disclosed or requires mandatory reporting. Not good. That's not just about data leakage, but again, any adverse impact reputation-wise is included there. And then again, we're really enforcing this by exception right now. So exceptions are requested in writing to my office. They are approved or not in writing. And that's not just the tool or where you're getting it from. That's also the use case and the data sets that you're using in that use case. So right now we're just collecting those. They're included in our ticketing system. you just reach out to the help desk in IT, talk about the tool you want to get, talk about why you want to use it and what for and which data you intend to feed from it or extract from it. And then we approve or deny those on a case-by-case basis right now. And then again, depending on the magnitude, we ask for additional details about roles and permissions. Who's going to do what? Who's going to make sure that this is being used safely? And then success criteria. Who's going to say whether or not this is effective and whether it's something we should continue to invest in? In addition to the technical controls, we've got a couple of operational and management controls. Existing business processes for budgeting and procurement say that I have to, my office has to review and accept anything. So that's technology contracts, any kind of new purchase, those all flow through my office. Enablement in terms of training. We just concluded our co-pilot flight school effort where we offered an hour-long training, basic training for co-pilot to everyone in LFUCG. In addition to that, we're going to be kind of creating a library of resources on the Internet where we capture use cases, where we offer other learning resources, that type of thing. This is my last slide. Would you like me to continue? Yes, please. Okay. So, again, you know, employees are our greatest resource here. Making sure that they're trained on appropriate uses of these products is something we want to deeply invest in. One of my favorite controls here is a guiding principles document that we've right now got in draft status, but it establishes sort of our foundational values to guide decision making. This policy is great, but it was released in October of 2025 and probably already needs some revisions. knowing that we've accepted the use of some tools here. So we're going to try to keep that evergreen, but it's really difficult. What the guiding principles do is establish just sort of our general ethos around how we're going to do this. And that's to say that any AI use in LFUCG will be safeguarded, transparent, human-centered, equitable, effective, and empowered, meaning that people will be trained on it. So that just, I think, will sort of establish just our approach in whole to what we intend to do with these things. And then we've got expert guidance from different organizations that myself and my team are members of. IONS is a cybersecurity faculty that really assisted with development of the policy. The Gov.ai Coalition is actually a coalition of, I think it's mostly city and county governments, but started in San Jose, California. Lots of my CIO colleagues in that group, and they've done a really good job of kind of dividing and conquering, consolidating materials, giving us templates to work with, and that sort of thing. Lexington is also, this is sort of a newish effort. Lexington is part of a 10-city cohort through the Knight Foundation and Harvard's Kennedy Center for Data Smart Cities called the Responsive Cities Network. We've just launched that last month. I think we've had one meeting with the other nine cities, but that'll be a three-year effort where we will, again, be sharing resources, sharing ideas, and hopefully getting some usable tools out of that effort. Gosh, I have one more slide. My apologies. These are our currently approved use cases. This, again, is not an exhaustive list. This is kind of the more recent ones, but you'll see the applications which have sort of been approved as AI use cases. Additionally, we are currently looking at some AI-assisted plan review and permitting technology, knowledge and voice agents for 311 or other resident requests. Language interpretation is in big, big, big demand, and there are a lot of tools that kind of focus on that use case. and then computer vision and object detection for multiple applications across government, lots in public works for, you know, detecting potholes in the street, managing traffic, things like that. So those are some of the additional use cases that we're actively looking at right now. Now we're on to questions. Apologies for going over. It's okay. Thank you, CIO Rogers, for that presentation. Colleagues, go ahead and sign in with your questions or comments. Council Member Savigny. Thank you, Chair, and thank you so much for the presentation. My question probably goes to probably the external, like our external partners. So I think we basically have cloud-based, I think, PeopleSoft right now, right? and we've got cloud-based camera systems through the police. Is our office, our CIO's office, are you managing all of those relationships to make sure that our external partners are somewhat compliant with the way that we want to do things? I don't know if our agreement with, think about it, If our agreement with Flock Camera basically states these are our policies, you have to adopt our policies, and internally you have to prove to us that you do it this way or something like that. So in general, we have given all of those vendors, and you're correct, they number in the probably hundreds by this point. The expectation is they will not introduce any new features without our consent. So by way of that agreement, I would expect them to not just turn on any AI capabilities. Now, that said, there are certain vendors we trust with and certain we watch a little more closely. But yeah, that just as an effect of our typical change management flow with those vendors, I expect that that would be disclosed, discussed, very well understood before they just implemented anything like that, particularly resident facing. And this will be included, the policy will be included in the list of things they're expected to comply with as we move forward with any renewals or anything like that. And is there anything that, as council members and council member offices, are there things that, for instance, when we deal with our legal staff, if it's something like we can basically say this is client attorney privileged, is there anything that we should be putting on our emails or even on our own Facebook? Anything that we should be doing with anything that's customer facing that might be more, that would be prudent and protective of what we're providing? Like is there, I don't know if they've come up with something that says, like this email cannot be used as data for artificial intelligence, if there's a standard disclaimer format or anything like that. So, no, not currently. Okay. That's not to say that we couldn't draft kind of a general purpose disclaimer with the assistance of the law department. It's interesting where requirements for disclosures have gone. When we first started drafting this policy, which is now about two and a half years ago, The expectation at that time was disclose what tool you use, disclose the source. And that has really, as use has evolved and been so heavily adopted, that has really kind of fallen by the wayside. You're not going to see footnote notations or anything. That's not to say that we shouldn't be clear about AI was used to produce this. but in terms of those disclosures, they've gotten way more minimal. So, yeah, I think that's something that, again, as leaders, we should be consulting with the experts on and agreeing to as standard practices. It's, you know, again, something hard to put in a policy because of the it depends factor. But I think it would be, you know, back in the interest of public trust, it would be good to be transparent about where those things are used. I don't know if that answers the question or not. I mean, it's a very tough area. So I appreciate you actually working on it. Thank you. That's all I have, Chair. Vice Mayor Wu. Thank you, Chair Wu. Thank you, CIO, for your work. And I appreciate us staying on top of this stuff because I think traditionally, especially governments have lagged behind technology and are always chasing in terms of our policy. How is our data stored? It depends on the data. So currently it is heavily siloed. We have data living in software applications, like our ERP system, you know, houses the data that pertains to our finances and our people. You know, we have data captured in personal files and shared files, just sort of the routine business. Those are right now stored on our premises, on our hardware, but my team is working through migrating those to a cloud, a secure, Microsoft secure cloud. I think they're maybe about 40% done with that project. And then we've got data contained in on-premise business applications still. Legistar is a good example of that. We've got data stored on paper and file cabinets. We've got data stored on big pieces of paper and file drawers. That answer is as complex as you might expect. Sure. And as you may also expect, the efficacy of any AI application is going to depend entirely on the accuracy, availability, and quality of that data. So, yeah, knowing where it is is sort of the first issue. I wanted to ask you about the co-pilot flight school, the training for this. Is this so far going to be voluntary training for any staff or employees that want to use it? Are any divisions or departments requiring their staff to get trained on it? I left that up to individual division directors. It was available to everyone outside. That excludes public safety, specifically police and E911. They're in a different tenant. They do not have this technology available, but for everyone else, it is the free version of Copilot called Copilot Chat is available already, and thusly the training was available to them as well. It's not mandatory across the board. Are we training supervisors on how to monitor compliance with our rules? That's actually one of the next things on the agenda, hopefully with the same partner who offered the basic literacy training, is that what I'm going to refer to as discernment training for leaders to be able to know this is a healthy, likely low-risk application of the technology versus something you should probably not consider doing. because, no, I don't think other than, you know, reading on their own that people are generally aware of the pitfalls of using this technology. So I would like to focus some learning on what that decision-making and risk management decision needs to look like. Okay. And I know all of this is new, especially our rules. We've had this policy in place for less than a year, right, since last October. have you all had any instances or experiences with compliance, with any issues that have come up so far with people using it, anything been flagged, anybody been disciplined, that sort of thing? Nothing severe. For most folks who had become devotees of the Church of Chat GPT, we were able to swap that out with a copilot license and say, off you go. So, you know, compare and contrast capabilities and let us know if this works for you. That's been successful most of the time. We did have a couple of folks who submitted use cases that I was not comfortable with, that I then discussed with our CAO and she was not comfortable with, and just things that we didn't think were the right place to make these investments. So, yeah, people have been very deferential in terms of wanting to do the right thing here. I have a couple more questions, but I'll sign back in. Thank you. Council Member Lynch. Thank you, Chair, and thank you, CIO, for the presentation. And my questions are kind of along the same lines as vice mayors regarding training. I took the training last week, and it was very, very helpful. And my first question was, do you all plan to offer that basic training again? Each of those sessions, we had seven, I think, eight sessions. Each was recorded, so we're going to kind of go through and curate the best recording out of all of them, you know, found our trainer on his best day, and make that available permanently on the Internet. and then, again, pair that with other resources like a prompt library. I would really love to have people share their use cases. You know, I used it for this and was very successful and create kind of a forum for that. So, yes, the answer is that basic literacy training, yes, will remain available permanently on the intranet. Wonderful. And then my second and last question, I was just curious how many internal staff took the training. Was it a great amount of folks? Was it a small percentage of all of our employees? So it was about 10%, which I'm trying to rationalize that with myself. It was about, I think, 324 was the exact number of people that we had enrolled. My expectation is that as folks start to see their colleagues utilizing these things, they'll hopefully say to themselves, well, that looks cool. I could definitely apply that to my work. and then they'll have the recorded training available to them. I really do think, and I've observed really in this organization for other things, it is very much a word of mouth kind of grassroots way that things like that take hold. So if you took the training and you learned to do something cool, share that with your colleagues, and hopefully we'll get everybody, you know, rising tide raises all ships. So we'll get everybody there. Thank you so much. Thank you, Chair. Vice Mayor Will. Thank you, Chair. You had mentioned the idea of some staff who are already using other AI like ChatGPT and kind of migrating them over. Does this policy apply to usage of other AI besides Copilot? This policy more or less prohibits use of other technologies besides Copilot. And it's because we don't have the enterprise versions of those tools to be able to do the data protection controls like we can do on Copilot. So, again, for Copilot, that's in our Microsoft tenant. We can create guardrails that make us breathe a little easier when it comes to the data being safe. With the commercially available tools, we don't have that same capability. And from what we've heard so far, and I don't know if this will remain the case or not, but from what we've heard so far, the folks who had kind of taken off using one of those commercially available tools and then pivoted into using Copilot, it's like for like. Copilot is really, it's the same back end as ChatGPT. It's the same model on the back end, just with the Microsoft controls and branding layered on top of it. I'm kind of curious about the idea of the prohibition of using other AI. I think about the work that we do now. I do some of it on Google Docs and Google Sheets. I use Google Calendar instead of Microsoft Calendar. Just kind of thinking through what makes sense and what's enforceable in a way, How would you be able to know if somebody produced a document with the help of Gemini or ChatGPT instead of co-pilot? So our visibility extends to our equipment, our accounts being your Lexington KWY.gov credentials, and our network. So you're correct. And if somebody just took data from here, took it home, produced something with another model, we don't have visibility into that. So this policy would not be enforceable. My hope is that, though, we are providing tooling that is capable enough to make that a less attractive option. And being responsible for a data leak is something that should be very scary. When I talk about that, I always make sure to say, personally, financially liable. That dissuades people from that. We do also manage by exception, too. If there is a reasonable use case for, you know, this tool operates much more capably over this one, that's something that I would welcome, that information. And then if, you know, if it's justifiable, create an exception for that use case. Okay. And then one of the examples that you used in terms of, like, use cases is redacting personal information from open records, right? Mm-hmm. One of the rules or the policies in this policy has to do with personal information. So can you kind of talk me through how if we're asking AI to go through an open records request, and some of them can be extensive, right, how are we asking AI to go through and redact personal information if we're also prohibiting the access to that personal information? From feeding it personal information. So it's another exception by way of that particular technology being adopted as our enterprise standard for management of open records requests. So it is allowed on that platform with that mechanism. And it is AI-enabled in as much as for those documents. it can detect something that looks like a social security number, something that looks like a credit card number, or an address, or a telephone number, and then still offers that up to the human reviewer to say, hey, would you like to redact these things? That confirmation still goes. So it's another area where, by exception, by making that the standard for that particular line of business, it's made okay because we're comfortable with the safeguards. Okay. I appreciate that. And I appreciate this policy being in place and this discussion. I've always said when it comes to technology, whether it's license plate readers or AI, I'm never afraid of technology or the tool. What's important to us is creating the guidelines, the safeguards, the accountability, who has access to it, usage and things like that. So as long as we have that kind of stuff in place, it makes me feel a lot more comfortable. Thank you, Chair. CIO Rogers, I just want to quickly say over 300 people taking a training sounds good to me. So I think for a first round, congratulations. I wouldn't shy away from that number because I think you're right that as word spreads, more people might take advantage of the training that you have recorded. I've sent the recording to a half a dozen people already this week. So, yeah, that's good. Council Member Curtis, do you have any closing remarks for this particular item? I don't believe we have any actions to take on this today. Yes, that's correct. So seeing as this is a CIO policy, there is not any sort of legislative action that this committee needs to take today. But I do have some remarks I'd like to give and also just to thank CIO Rogers again for her work on this issue and her team's work on this issue and for her willingness to be here and give us such a thoughtful and in-depth presentation today. And I want to thank you guys for having these questions and insights and a willingness to start having a conversation about this technology and its usage within our government. And that's what I believe today has been, which is an overdue start on this body to a necessary conversation. I look forward to working with you all in the years to come to ensure that both our CIO policy stays up to date as AI continues to evolve. and also to take on some other applications. And I want to be clear that when I am saying this, I'm speaking for myself and myself alone. But I hope that we will not lack the political willpower to have difficult conversations about other potential threats that AI might pose to our residents and to our city. Our constituents have legitimate concerns about the rapid proliferation of data centers across the Commonwealth in our country. I share those concerns, and that is a part of the conversation that we are anticipating having in the near-term future. And it's my hope that we will work to protect Fayette County's land and resources in that process. And our constituents do also have legitimate concerns, as Councilmember Civigni pointed out, about the utilization of AI and policing, including by private companies that LFUCG has contracts with, like FLOC. I share those concerns, and I believe that this council should be taking the matter seriously and working to address it with the urgency it deserves, even if today is not the appropriate forum for that. And I do share my constituents' concerns about our willingness to actually have those conversations in a way that this moment demands and to make the right decisions even when it's hard. So I hope we can prove them wrong and show that we're up to the task. Thank you. Thank you. And, colleagues, I'll give a brief update on where our data center conversations is at the end of our agenda when we're talking about items in committee. Our next agenda is a presentation on the sustainability standards for LFUCG buildings. Colleagues, you may recall that due to our discussions around environmental sustainability for government spaces, we've talked about it within Environmental Quality and Public Works when we've had James Bush come and present to us on our Energy Initiatives Fund and then more recently with the City Hall discussions. I have asked our Environmental Commission to advise us on standards for our LFUCG spaces. We have set community goals with our comprehensive plan, with the Empower Lexington plan, and our Race to Net Zero plan. So it's important for LFUCG to be leading by example on the expectations we set for the community as a whole. And often times what I have found is that the conversations around sustainability when we are building new construction or doing renovations is that the sustainability piece of it is seen as an add-on or incorporated depending on resource availability. However, environmental sustainability is also a cost-saving measure. And it should be an essential part of our design conversations from the very beginning. So I'm grateful for the Environmental Commission for their work on the white paper and the sample ordinance that are in our packet, and I'm looking forward to the presentation and discussion today. So I'm going to turn it over to Patrick Mason. He is the chair of the policy committee for the environmental commission and we have a couple other if you want to wave environmental commission members here today to support this work so i'll turn it over to you thank you thank you chair and thank you to the committee for hosting us today it is a great honor to represent the environmental commission which is of course volunteers who've poured a lot of work into this ordinance proposal that you have before you today so i especially want to recognize our chair Blaine Early as well as our vice chair Kathy Clement for their work on this as well as the rest of the environmental commission. So to begin this proposal is charting a clear and legally durable path forward beyond the aspirational goals that council member Sheehan just mentioned. This is our vehicle to operationalize Lexington's target of net zero emissions by 2050. As you may know, we are in Kentucky, and so a lot of our energy is from fossil fuels. They are not from renewables. And up until now, we have had some aspirational resolutions and intents to do better, to do better as Lexington and as a government. But those aspirations don't really change how our buildings are built, and they don't change some of our budget priorities. So this ordinance proposal codifies our intentions into enforceable law, something that goes beyond administration to administration, and something that our staff in LFUCG can point to when they look at new buildings or renovations. And so this is rooted, like I said, in some of our established council past goals, especially our comprehensive plan, Theme B, points directly towards getting to those climate goals by 2050. And trying to meet those goals makes us an example, not just in Kentucky, but across the country. And we can't really ask our municipal stakeholders, our commercial tenants, to do better about energy unless we're starting as a municipality ourselves. So this ordinance proposal, it doesn't reinvent the wheel. If you're into intercity rivalries, Cincinnati and Louisville have already started. They are the reason why we would not be a guinea pig in this situation. Cincinnati uses mandatory benchmarking and building performance standards. I'll get into those definitions in a minute. to help catalyze some of their own efficiency. And Louisville leverages an energy innovation fund to make sure that they have those high return investments in their retrofits on city buildings. So by adopting standards similar to them, we're kind of competitive in our own region. And just to note on the building performance standards, they're not so much prescriptive. It's not about what you must install. It really leaves it up to the experts. So it's performance-based. It says we have the ability to innovate and reach some of those energy goals. To get into a little bit of the specifics on how Louisville fared, They've so far had $3.2 million in cost avoidance just by looking at some of the accounts that they were using to track utility costs. And they've also had a 7.5 portfolio energy reduction since 2019. And in Cincinnati, they've had a 100% project cost recovery through their verified ongoing operational energy savings. Not to say that that will necessarily be the case for us. This proposal is meant to be foundational across government. Like I mentioned, this is durable policy, not just projects. Now, we have a few great projects that we've done in the past that sort of are guiding light for this ordinance proposal. Our fire stations, several of them already have solar panels installed on rooftops. and then look no further than Locust Trace Agri-Science Center for their net zero attainment. Now, that's not to say that this ordinance is making us go completely smart buildings, but it definitely gets us on the right path. It gives us something to strive for in a black-and-white way. And it gives us not just a year-by-year look at these savings and these efficiencies, we're looking at a 30-year return on investments for some of our major systems. So we've set some goals. 20% reduction in energy use intensity by 2035. And looking further out, our goal is a 100% reduction in on-site fossil fuel use by 2045. So this looks at gas furnaces, boilers, not the broader utility grid. And for resilience, we're also looking at a 30% goal in efficiency above the current state code, which happened to be passed in 2012. So by sort of measuring ourselves up against that, we have a benchmark that Energy Star can track and give our energy manager a platform to work toward. So what are building performance standards? Normally, city codes, you know, these building codes are very prescriptive, and they tell you exactly what you need. A building performance standard is sort of like how well we are actually meeting these energy use intensity goals. It's basically how, basically all of these, taking an energy audit or report, maybe annually, to give us some transparency into, you know, why have heat pumps removed? And why have them switched out for something higher efficiency? You know, why replace those outdated boiler systems? Because we can actually see the measures over time. And, of course, it allows for new technologies as they emerge. Additionally, our ordinance includes a solar readiness provision. So this is really putting our money where our mouth is in terms of solar energy, in terms of getting it on rooftops, and future-proofing some of our assets to be smart from day one, not just looking at overtime. And especially when you're looking at roof construction, you've got to get the angles just right for solar. So, again, looking at Fire Station 21 and Glen Eagles, we could save thousands of dollars by having these cover our rooftops. Just like when we point to some of those big buildings in downtown and we say, why is there not solar there? Well, we can see now city buildings taking that on, as well as having room for battery storage to sustain those solar savings. So operationally, we've built in some fiscal wisdom here, if I might say so. We already have a city energy manager. We think the city energy manager does a great job, so this is not a new position. But it gives the city energy manager the tools necessary to be even more successful and to sort of standardize those savings that we're looking for over time. It's a codification of some of those priorities that we have. And it also introduces a dedicated revolving fund so that we're not just looking for money in the couch cushions when we're ready to say, hey, we want some new windows on this building. What is it going to cost? We look from the very beginning and say, these are the kinds of proposals that we want, and this is what we're expecting as government. That's why we also included contractor incentives as a requirement that we get proposals from day one for some high efficiency renewable alternatives during the planning process. Because retrofits can be entirely too expensive when you're looking 10, 15 years down the road. and finally this is sort of an overview of of the scope and and what we're looking at in terms of of the included buildings and we also want this policy to be effective upon passage this will of course give the city time to prepare internally as well as start measuring some of those goals, report that to council, and get us where we need to be in terms of our efficiency. With that, I would love to take any questions you may have. Thank you, Mr. Mason, for the presentation. Colleagues, please sign in with your comments or questions. We do have, obviously, Mr. Mason is here for questions about how the Environmental Commission came up with their recommendations, but we also have folks, Commissioner Ford is in the back and his staff from General Services, and we have James Bush here from Environmental Services and other folks. So if we have questions that come up where they might be able to guide us or give input, they are present today as well. So go ahead and sign in. Vice Mayor Wu. Thank you, Chair. Thank you, Mr. Mason, for the work that you're doing on this. When thinking about the goals set forth in this ordinance, and I know sometimes when we think about setting goals, you know, do we set one that's just a little bit out of reach to kind of motivate us to charge forward? Do we set one that's a little bit lower that we know we can hit and then hopefully go for stretch goals? In your mind, where are these goals set in that sort of range? These goals are designed to be ones that we can meet. We have already looked to other cities who've been able to do this over the course of several decades. It gives us about 30 years to make some measurable achievements on some of these energy efficiencies. And think about it this way, whenever we have something break or the HVAC goes out, that is an opportunity for us to make a better choice. And those usually have maybe a tent, depending on commercial systems, but those can only last so long. So these are attainable goals. Thank you for that. Is the intention also to think more about new builds and things that we're creating along the way to kind of bake these sustainability elements, energy elements into it? Or how much of it is targeted towards existing buildings and trying to retrofit them to get them to a particular standard? Well, it won't cause any immediate retrofits. These are as we go, basically. Of course, if we wanted to build something into the plan, that is certainly council's prerogative. If we wanted to prioritize new retrofits on older buildings or existing buildings. But in terms of those new purchases, new constructions, that would have the most immediate impact as well as the most cost savings. Yeah. And I don't know if there's necessarily a satisfactory answer to this or if you've gleaned anything from Cincinnati, Louisville or other comparable cities. What kind of teeth are built in or can be built in to sort of keep us accountable, right? When we set goals and then if we don't hit them, what then? You all are the teeth. So basically, we have set in an annual reporting mechanism from our city energy manager to say, hey, this is what I'm seeing. This is what's happening with our portfolio right now. This is where we are missing some spots. And if we are not seeing some of those achievements over time, it would be incumbent upon council or upon the administration to say, why is that? Why are we not making, are we making, you know, is our revolving fund fully vested with enough to make it go as far as we want? So I think that those annual reports give us pause for question and then allows council to continue to reaffirm those priorities. Let me ask you about the revolving fund. Is the intention or the idea behind it to seed it from one particular year's budget and then build it in so that it is something that has money coming back to it and kind of self-perpetuate? Or are we talking about a specific tax that goes directly to this fund like a lot of our very specific funds? Great question. So some comparable cities, not just Louisville and Cincinnati, they did seed it with a few hundred thousand dollars and went on from there and went on implementing their program. Others actually, they started looking at where we are actually not accounting for some of our energy savings and counting those as savings toward new construction. So whatever we saved on a budget for this building, well, now we can put that back towards our revolving fund. So it could go either way. There has been success, though, with going ahead and getting it started, but allowing our energy staff to say, yes, we've made these savings, it's going into this fund, and we can appreciate it from there. Okay, thank you for that. Thank you, Chair. Councilmember Savigny. Thank you, Chair, and thanks so much for the presentation and the Environmental Commission's work on this. I do appreciate it. My question might be a little bit more geared towards our staff on this one, but I'm thinking, what I'm kind of thinking for us to understand what we're biting off here, off here because i i do think it's it's all very important and i do think um i would i would love to adopt an ordinance like this um i do think it's important for for us to understand as a body and for the administration to understand um as an organization the the bite that they're we're taking off to commit to something like this so to me that would typically be there would be some sort of gap analysis between here's all of our buildings you know here's the issues here's the gaps in those buildings we assume that um we would retrofit this building we would probably move out of this building we would build a new x facility that would do this i mean there's there's just all these different moving parts that um my guess is the scale of that is something that this body would need to understand better. And I don't know if you probably didn't do any of that work. You're focused mostly on the policy piece of it. Is that correct? So we did do some work on the we did consult with some of the subject matter experts on this. And it is attainable. There would need to be a plan, of course, to implement it. One of the first parts of that plan would be to see where we're at now and see doing some of that benchmarking work. Where are we starting from? Where do we need to be to be 30% above the current energy code? And so I definitely welcome the opportunity to see what kind of regulations would need to be in place to make sure that happens and make sure that that's attainable. But I appreciate any of council's resources to operationalize that. Yeah. I also think that there's probably a strategy when we're dealing with things like solar energy that since we've gone into an agreement that we're leasing our landfill, it was probably one of our biggest land masses that we could do something of scale at. and I do know when it comes to clean energy, the scale matters. So I think that's just something that would make us rethink that agreement a little bit maybe to see if we would funnel more energy, we would funnel energy off of that. I really do appreciate the work. It creates a lot of questions, and sometimes I think that's what ordinances do is it forces you to think and it forces us to take this, kind of go through law, deal with our folks here, and just say, how do we make an ordinance that has enough teeth in it so that newer councils as they come down the road aren't just going to abandon this, things like that. So thank you, Chair. Thank you for the work. And just to quickly comment, I asked the Environmental Commission to give us what they felt like was their ideal standard. And then I have already met with General Services and Commissioner Ford, and I'll talk a little bit at the end about next steps in that conversation to kind of bring it all together. Council Member Curtis. Thank you, Chair, and thank you, Mr. Mason, for being here. It's good to see you in Council Chambers. I want to just very briefly express my gratitude for this proposal, this policy, and this presentation. I don't think it's any secret to anybody here that I have felt a great deal of frustration with what I feel has been our government's willingness to pass policies or sloganize the climate crisis to not actually tackle into it tangibly and try to be aggressive in preventing what we can prevent from the coming climate crisis. crisis and I think that this policy is a very tangible step forward. It gives us some very good teeth to actually start attacking that and it gives us an opportunity to not just blame our residents but to actually lead by example on this. So I'm very appreciative and I'm looking forward to supporting it. Thank you. Councilmember Reynolds. Thank you, Chair. And thank you, Mr. Mason. And I just wanted to say, first of all, that I really appreciate the Environmental Commission's work and setting forward these standards and that we need things like this to help us move the needle and move forward in a positive direction. So I appreciate it and I'm in agreement. I just wanted to ask our staff, maybe Commissioner Ford and those in the back row, how did you feel, do you feel that these are standards that are practical that can be worked with and attained? or do you feel like there needs to be a few things adjusted to make it something that is possible? I just wanted to hear your feelings on that. Thank you, Council Member, and good afternoon, Committee. As Council Member Sheehan mentioned, she has met with us. We envision meeting continually moving forward. We think that the work of the Environmental Commission is aspirational, and I think it's going in line with the goals that our community have. As with all things that are tough, collaboration, coordination, alignment, bringing minds together to figure out the best thing. One thing that we would caution is, and I think everybody here would agree, well-intended policy that might have unintended consequences. That's why this interim period of study and collaboration and working together might ultimately present a good outcome for all. That may take a little time, but our staff is open. We have kindred agencies that we work with, of course, James and Jennifer from EQPW are here. And so that's our thought. There's opportunity. There's opportunity to work together to come up with something that is tangible that can be done moving forward. If it's the will of the council and the commission to bring forth policy, our staff stands ready to work with you guys to find the very best policy that can be implemented. Okay, thank you. I really appreciate that. Thank you, Chair. Thank you again for the Environmental Commission and their guidance. We don't have an action to take on this today, colleagues, but as I mentioned, I plan to take this policy and these recommendations to our LFUCG staff, including the commissioner's office, and then having our environmental services folks weigh in, and having our environmental commissioners as a part of those meetings so that we can take these recommendations and look at how we can adopt them into policy for our LFUCG government spaces. So we'll work on that, and then I will be bringing this back to the committee for final approval. So thank you again. Thank you. This brings us to our last agenda item. Colleagues, sign in if you have any updates or changes to items on the referral list while I'm looking for logins. Because we just mentioned in our earlier AI presentation data centers, and this has been a pretty energetic conversation across Kentucky, I do want to let you know that the Blue Sky Small Area Plan is on our agenda for our next meeting in July. And in their last work session in May, the Planning Commission heard a presentation on the first draft of that small area plan. And the consultant working with LFUCG on the Blue Sky Plan is recommending some updates to our industrial code as a part of that overall recommendations there that would include updating zones before I-1 and I-2. And some of those revisions do pertain to data centers. So if you want to see some of that work, you can look at the last Planning Commission work session. But the Division of Planning is working towards the completion of that small area plan. and then we'll have the presentation in July and then move from there. But just to let you know, that conversation is moving forward and we will have updates as they are ready. I don't see anyone else assigned in for items referred to committee. So that is going to conclude our agenda. Our next meeting is on July 1st, which is a Wednesday. So just be careful on your calendars because July is a little bit different. and I will call our meeting adjourned. Thank you, everyone.