The Complete Guide to Using AI in the Government Industry in Fort Collins in 2025

AI matters for Fort Collins government in 2025 because state and federal guidance now expects careful, mission‑aligned adoption: the Colorado Guide to Artificial Intelligence frames GenAI work around three pillars - governance, innovation, and education - and requires OIT risk assessments for all GenAI efforts, including vendor projects (Colorado Guide to Artificial Intelligence (Colorado state AI guidance)); the federal GSA AI Guide for Government (federal AI guidance on trustworthy AI deployments) adds practical advice on organizing teams, data governance, and lifecycle testing for trustworthy deployments.

For Fort Collins that means starting small with pilots that map directly to the 2024 Strategic Plan's organizational priorities and measurable outcomes, while training staff in workplace AI skills - an approach mirrored by practical courses such as Nucamp's AI Essentials for Work syllabus (practical AI skills and promptcraft for non-technical public servants) that teach promptcraft and operational use cases for non‑technical public servants.

Understanding 2025's AI breakthroughs means recognizing that Fort Collins must move from curiosity to practical, governed action: national and vendor moves - such as AWS's announced $30 billion AI infrastructure investment - have made scalable cloud + GenAI stacks available, but public agencies still face legacy systems, limited budgets, and skill gaps; Presidio's field work highlights four practical trends that matter locally (closing readiness gaps, mission‑ready vertical solutions, people‑centric responsible AI, and budget‑focused sustainment) and notes that approaches like Human‑AI (HAI) have helped some agencies off modernization timelines, a tangible win for city IT planning (Presidio GenAI trends reshaping the public sector in 2025).

State process and oversight are converging too - the Colorado Artificial Intelligence Impact Task Force set formal reporting obligations for 2025 that Fort Collins projects must respect (Colorado Artificial Intelligence Impact Task Force 2025 page) - and the federal Administration's AI Action Plan frames policy priorities that should shape local procurement, workforce training, and risk assessments (Federal Administration AI Action Plan (July 31, 2025)).

The practical takeaway: prioritize small, measurable pilots that target one city priority, apply tested modernization methods to cut timelines, and bake governance and workforce skilling into every pilot so results scale without surprise costs.

shave a year or more

In 2025 the most popular AI tools for public‑sector pilots are the same cloud‑and‑model stacks that dominate enterprise: Microsoft Azure and Azure OpenAI for integrated copilots, AWS (SageMaker/Bedrock) for flexible foundation‑model access, Google Vertex AI and Google AI Studio for fast prototyping (including free tiers and Gemini access), plus GPU‑heavy infrastructure from NVIDIA for any local training or large inference loads; municipalities planning pilots in Fort Collins should prioritize cloud‑hosted, managed services to lower up‑front costs and use Google AI Studio's free tiers to prototype before committing to paid compute, because NVIDIA today controls ~92% of the data‑center GPU market - a concentration that affects pricing and procurement decisions and makes early vendor‑agnostic architecture important (NVIDIA, Microsoft, and AWS market overview, Google AI Studio free tiers and Gemini overview).

“We're finding tangible ways to leverage GenAI to improve the customer, member, and associate experience. We're leveraging data and LLMs from others and building our own.” - Doug McMillon, CEO of Walmart

Fort Collins must treat AI projects as regulated programs: Colorado's statewide GenAI policy requires that all generative AI efforts - including third‑party vendor solutions - go through the Governor's Office of Information Technology (OIT) for intake and a NIST‑based risk assessment, so local pilots align with the state's governance, innovation, and education pillars (Colorado Guide to Artificial Intelligence).

The state's GenAI Risk Index and guidance spell out prohibited activities (e.g., facilitating illegal acts or undisclosed use of GenAI), high‑risk uses (e.g., evaluations of people, handling CJIS/PHI/PII, production code) and medium‑risk scenarios, and OIT expects procurement terms and vendor contracts to protect data and comply with SB24‑205 (GenAI Risks & Considerations - risk index and prohibited/high‑risk uses).

Practically, staff may not use the free ChatGPT on state devices due to contract conflicts with state law, and Fort Collins should plan pilots around OIT's intake, training, and attestations - Colorado's 90‑day Gemini pilot offers a tested model for structured evaluation and rollout (State guidance: Free ChatGPT Prohibited).

“Gemini has saved me so much time that I was spending in my workday, doing tasks that were not using my skills. Since having Gemini, I have been able to focus on creative thinking, planning and implementing of ideas - I have been quicker to take action and to finish projects that would have otherwise taken me double the time.”

Cybersecurity is the top near‑term threat to Fort Collins' critical infrastructure, and local water utilities should treat AI pilots like any other operational control: federal rules already require community water systems serving 3,300+ people to consider cyber threats in their risk and resilience assessment and emergency plans, so Fort Collins must use the AWWA Cybersecurity Assessment Tool & Guidance to map controls to the NIST Cybersecurity Framework, run the Getting Started Guide and templates (Risk Management Plan, Incident Response Plan), and prioritize Cyber‑Informed Engineering for process‑level defenses (AWWA Cybersecurity Assessment Tool & Guidance for Water Utilities).

At the same time, AWWA's AI content policy warns not to upload sensitive or customer data to public AI services and requires clear oversight and attribution when AI assists content - practical rules that translate into procurement clauses and staff attestations for Fort Collins pilots (AWWA policy on responsible AI tool use for content production).

The immediate, actionable step: run the AWWA assessment, document prioritized controls tied to NIST/AWIA §2013, schedule a tabletop exercise within 90 days, and embed “no sensitive data to public LLMs” into vendor contracts so AI pilots improve service delivery without exposing customer PII or process‑control systems.

“The policy balances innovation with ethical considerations and protects against reputational damage while reinforcing AWWA's credibility.” - Mike Hiskey, AWWA Information Technology Director

The spring 2025 CSU hackathon in Fort Collins, co-led by Colorado State University and federal partners, turned classroom learning into mission-ready prototypes: over 100 CSU students on 20 teams used Forest Service satellite and LiDAR data and AWS SageMaker to build machine‑learning models that automate mapping of uncharted National Forest System roads and design rover navigation algorithms for rugged‑terrain rescue scenarios, outcomes that USDA says will modernize road maps to improve fire response and public safety while feeding NASA robotics efforts (USDA recap of the CSU hackathon).

The event - hosted by CSU's DevNet and the Career Center - paired technical mentorship from USDA, NASA, and AWS with career panels and low‑cost prototyping that produced tangible deliverables (winning teams received recommendation letters from USDA and NASA), creating a repeatable model for Fort Collins to tap university partnerships for rapid, risk‑reduced AI pilots that both solve local service challenges and build a direct recruitment pipeline into public tech roles (CSU coverage of the hackathon, NASA summary of student approaches).

“The students picked it all up quickly, implementing techniques like transfer learning, data augmentation, and resampling that allowed them to come to a viable solution. I'm looking forward to digging into the results more and evaluating how they can help improve our work in delineating an accurate representation of the FS road network.”

Begin with a tightly scoped, auditable plan: 1) run an AI readiness assessment across opportunity discovery, data, IT/security, risk/governance, and adoption to pick one measurable pilot (ICMA's “AI Readiness Assessment Checklist” is the practical template); 2) inventory and classify datasets, then submit the use case to Colorado OIT for the required GenAI risk assessment and intake so state rules and the Statewide GenAI Policy are followed (ICMA AI readiness assessment checklist for local governments, State of Colorado OIT guide to artificial intelligence and GenAI intake); 3) choose a cloud sandbox or FedRAMP‑authorized pilot environment and run a 90‑day controlled sandbox (the state's Gemini pilot is a tested template) to validate outputs before production; 4) create simple governance - assign an AI steward, establish review gates, and require vendor attestations that forbid uploading sensitive data to public LLMs; 5) harden controls for critical systems (for water and utilities, run the AWWA Cybersecurity Assessment Tool, map controls to NIST/AWIA §2013, and schedule a tabletop within 90 days); and 6) upskill staff with role‑based training and document acceptance criteria so the pilot yields a vendor‑ready contract and a validated MVP. The so‑what: one focused 90‑day pilot tied to a single city process (for example, permit intake automation) can produce a tested MVP and procurement-ready requirements in months - turning uncertainty into a governed, scalable pathway for Fort Collins.

Fort Collins' workforce pipeline already has clear, local pathways into AI roles: Colorado State University's undergraduate Computer Science major with an Artificial Intelligence & Machine Learning concentration pairs a rigorous 120‑credit CS degree with targeted AI/ML coursework and capstones that prepare students for entry technical roles (Colorado State University Computer Science major with AI/ML concentration); for rapid upskilling, CSU's Master of Applied Statistics - Data Science Specialization lets full‑time students finish in one year (up to 30 credits) with applied machine‑learning and statistical computing courses aimed at making graduates job‑ready for government data roles (CSU Master of Applied Statistics - Data Science Specialization); for working professionals and remote learners, Colorado State University Global offers an accredited online Master of Science in Artificial Intelligence and Machine Learning with courses in programming, algorithms, computer vision, and ML to bridge practical skills gaps (Colorado State University Global online MS in AI & Machine Learning).

Complement these degree routes with short, applied training (fast Python/ML bootcamps and a 260‑hour Data Science & AI course option) so city hires can reach operational competency quickly - the so‑what: a one‑year MAS or a targeted online master plus a short bootcamp can convert local talent into deployable analysts and ML engineers in months, shrinking hiring lead times for Fort Collins government projects.

Fort Collins can seed practical, low‑cost AI pilots by partnering with Colorado State University to run short hackathons and internship projects that turn student prototypes into procurement‑ready pilots: at a March 2025 CSU hackathon more than 100 students across 20 teams spent 24 hours using Forest Service satellite imagery and AWS tooling to build ML models that automate mapping of uncharted National Forest System roads and pathfinding for rover navigation - outcomes USDA says will modernize road maps to improve fire response and public safety (USDA recap of the CSU hackathon mapping future).

Complement those sprints with semester‑long internships at CSU's Geospatial Centroid to produce shareable GIS Story Maps (for example, Fort Collins' paved trails history) and staffed handoffs that convert prototypes into maintainable data products and hiring pipelines (CSU Geospatial Centroid student projects and GIS story maps); the so‑what: prototypes from short events can deliver targeted capabilities - improved road maps and trail visualizations - that materially cut response times and inform capital planning for trails, parks, and wildfire mitigation.

“The students picked it all up quickly, implementing techniques like transfer learning, data augmentation, and resampling that allowed them to come to a viable solution. I'm looking forward to digging into the results more and evaluating how they can help improve our work in delineating an accurate representation of the FS road network.”

Fort Collins' responsible AI roadmap in 2025 is straightforward: treat each project as a regulated program - start with the state OIT intake and NIST‑based risk assessment, run a 90‑day controlled sandbox tied to one measurable city priority, assign an AI steward with clear review gates and vendor attestations (including “no sensitive data to public LLMs”), harden critical systems with the AWWA cybersecurity assessment for water utilities, and align governance with federal best practices such as the GSA AI compliance plan so oversight, risk adjudication, and continuous monitoring are baked into every stage; practical supports include the Colorado OIT GenAI guide for intake and risk rules (Colorado OIT GenAI guide for government intake), the GSA AI compliance plan for governance bodies and safety teams (GSA AI compliance plan for federal AI governance), and focused workforce upskilling like Nucamp's AI Essentials for Work syllabus to get non‑technical staff productive in promptcraft and use‑case validation (Nucamp AI Essentials for Work syllabus).

The so‑what: one focused 90‑day pilot tied to a single city process can produce a tested MVP and procurement‑ready requirements in months, turning uncertainty into a governed, scalable pathway for Fort Collins.

“Artificial Intelligence is rapidly transforming how we live, work, and govern.” - Governor Katie Hobbs