Top 10 AI Prompts and Use Cases and in the Government Industry in Puerto Rico

Puerto Rico is at a pivotal moment for public‑sector AI: V2A Consulting's 2024 state survey shows 84% of local organizations (and 94% of multinationals on the Island) have applied AI to at least one business function, with marketing and service operations leading adoption and administrative automation emerging as a clear win for governments; yet the rollout faces real hurdles - 59% cite a lack of in‑house expertise and experts at Tech Day Puerto Rico warned of infrastructure and regulatory gaps.

Legislative momentum is building too, with proposed bills to create an AI Officer and advisory council to guide safe deployments, and community groups highlight AI's role in planning and resilience amid the “great tsunami” of change.

Closing the skills gap matters: practical programs like Nucamp's Nucamp AI Essentials for Work bootcamp (AI Essentials for Work syllabus) can help public servants turn pilot projects into trustworthy, operational services while keeping transparency front and center; see the full V2A survey V2A Consulting Puerto Rico AI survey 2024 and Foundation for Puerto Rico's planning perspectives Foundation for Puerto Rico resilience with artificial intelligence.

“A significant 84% of local organizations report having applied AI in at least one business function. More importantly, results suggest that AI is starting to deliver value to Puerto Rican organizations.”

The methodology for identifying Puerto Rico's Top 10 AI use cases married legal‑AI benchmarking practices with pragmatic vendor comparisons and local compliance filters: a task‑typology approach (Find, Learn/Investigate, Create/Summarize) drawn from AI law librarian benchmarking and evaluation guides, prompt‑engineering rubrics for reproducible testing, and vendor feature/security checks that matter to government adopters.

Benchmarks prioritized citation integrity and administrative‑code fidelity (important for municipal ordinances and procurement reviews), using vendor feature sets and pricing as tie‑breakers - examples include Lexis+ AI's document and code‑compare capabilities and the platform tradeoffs shown between Gemini and Copilot in independent comparisons - while flagging FedRAMP/ISO‑style controls and data‑sovereignty concerns highlighted in government guidance.

Reproducible prompts and multi‑model checks were emphasized (the legal librarian community's “run the same hypo across AIs” practice is a vivid reminder: one benchmark asked a font‑copyright question across seven systems to expose citation drift), and each candidate use case was scored on usefulness to Puerto Rican workflows, privacy/compliance fit, and ease of staff reskilling so local teams can move pilots into production with lower risk; see the AI Law Librarians benchmarking resources and LexisNexis notes on AI code compare for the administrative‑law workstreams that informed selection.

“I don't trust standard AI for any legal work, but with Genie, I have a lot of confidence! I love the interface, how easy it is to use and all the many improvements that have been made since I first started using it.”

Administrative automation in Puerto Rico's public sector can get an immediate productivity boost by putting Microsoft 365 Copilot to work - automating meeting notes, surfacing who said what, and turning chat and transcript threads into assigned action items so

“volunteer note‑takers”

become a thing of the past; see Microsoft's guidance on how Copilot captures and summarizes meeting outcomes and exportable recaps (Microsoft Copilot in Teams meetings: capture and summarize meeting outcomes).

Practical prompts (goal + context + expectations + source) help clerks and municipal staff get reliable outputs - e.g.,

“Write a summary based on all emails from [X] in the past two weeks”

- and the prompt gallery shows common templates for agendas, training outlines, and follow‑up emails (Microsoft Copilot prompt gallery: templates for agendas, training outlines, and follow-up emails).

Important guardrails for island agencies include licensing and transcription settings (Copilot requires appropriate Copilot/Microsoft 365 licenses and meeting transcription to enable full recap features) and the simple but crucial step of verifying AI outputs before publishing - so Copilot speeds paperwork while keeping compliance and auditability intact (How to generate meeting notes with Microsoft Copilot).

For the Puerto Rico Emergency Management Agency (AEME‑PR), Gemini-powered constituent communications can close the gap between technical guidance and people on the ground by turning official, multilingual preparedness content into clear, timely alerts and short action items: automatically generating concise Spanish summaries of NOAA hurricane preparedness guidance (NOAA hurricane preparedness guidance) and Ready.gov emergency preparedness materials (Ready.gov emergency preparedness materials), surfacing the Ready Family Communication Plan fillable card in Spanish (Ready.gov Family Communication Plan (fillable card in Spanish)), and pointing residents to the Listo Puerto Rico emergency preparedness site (Listo Puerto Rico emergency preparedness site) or the FEMA Mobile App for shelter and recovery details (FEMA Mobile App for shelter and recovery information).

Because Ready highlights translations (Ready.gov Spanish translations) and the CDC provides parallel English/En Español post-storm health and safety guidance (CDC post-storm health and safety guidance and recovery checklists), AEME‑PR workflows should pair model outputs with source links and verification steps so every alert includes a trusted reference (for example, the Spanish "Are You Ready?" guide on Ready.gov (Spanish "Are You Ready?" guide) or the CDC hurricane recovery checklists (CDC hurricane recovery checklists)) rather than standalone advice.

The result is practical and human‑centered: short, bilingual messages that respect limited attention during outages, route people to authoritative resources, and make it easy for municipal call centers to escalate questions with context-rich summaries from NOAA and Ready.

For San Juan's permit portals, deploying Azure OpenAI and Microsoft Power Virtual Agents should begin with old‑fashioned design work: sketch the decision tree on paper and keep the bot focused on one task at a time - permit status, fee estimates, or required documents - so users aren't trapped in a Rube‑Goldberg flow; Tidio's chatbot flowchart guide shows why a seven‑question trunk can explode into 128 scenarios if not pruned and why “one decision bot = one topic” speeds resolution (Tidio chatbot flowchart and decision tree examples).

Follow Comm100's journey‑mapping advice to map trunks, branches, and escape routes for ambiguous inputs, add clear button‑based choices for low‑bandwidth mobile users, and always surface a bilingual live‑agent handoff when the bot can't resolve a case (Comm100 chatbot decision-tree journey mapping guide).

Finally, bake compliance and hosting choices into the design from day one - FedRAMP/ISO security expectations matter for public portals and are worth flagging early in vendor selection (FedRAMP and ISO security compliance for government portals) - so the result is a fast, bilingual self‑service channel that hands off smoothly when human judgment is required and avoids dead‑end frustration.

“Chatbots are only as good as the narrative itself, and storytelling is way more important than cutting edge tech or continuous content sheets.”

When Bayamón's legal team needs to modernize ordinances or reconcile municipal codes with island‑wide statutes, a grounded legal AI workflow can cut heavy lifting without cutting corners: Lexis+ AI legal drafting and research (LexisNexis) offers a private Protégé workspace to draft, iterate, and “summarize in moments,” pull internal precedents via DMS integration, and Shepardize® citations so every clause links back to authoritative law; see the Lexis+ AI legal drafting and research page for capabilities and security notes.

Paired with a Word‑integrated drafting layer like Lexis Create+ drafting tool for Microsoft 365 - which populates tables of authorities, redacts, and inserts proven clause language - municipal drafters can produce a clean, compliant ordinance draft and an event timeline from a pile of meeting notes and filings, all while keeping documents in an encrypted Protégé Vault.

For context on statutory powers that matter to local rulemaking, cross‑checking drafts against Puerto Rico statutes Title 9 §2004 - municipal authority and procedural requirements helps ensure ordinance language aligns with territorial authority and procedural requirements, turning manual citation hunts into auditable, verifiable drafts that make legal review and public transparency far more manageable.

When Puerto Rico procurement teams and records officers face a stack of RFPs, bids, contracts and FOIA requests, LexisNexis Protégé inside Lexis+ AI offers a secure, workflow‑oriented option: upload document collections into a private Protégé Vault, generate concise AI summaries, create a graphical timeline of procurement events, and surface citation checks and precedents so reviewers can spot inconsistencies faster and build an auditable response package for public records requests.

Protégé's DMS integrations and personalized profiles mean municipal counsel can ground outputs in local ordinances and internal files while keeping data encrypted and privacy controls intact; paired with Lexis+ AI's Shepardize® citation validation and Vault‑based secure databases, the tool reduces repetitive chasing of documents and turns a pile of paper into a single, searchable workstream.

For government adopters, embedding these processes alongside FedRAMP/ISO security planning is critical - see the LexisNexis Protégé overview and the Lexis+ AI legal AI page for capabilities and security notes, and consult local guidance on FedRAMP and ISO expectations for public portals.

“[This] marks a step change in our legal AI functionality, whether legal professionals are using their own internal data or LexisNexis trusted resources.”

Puerto Rico's Governor's Office can shorten the time between storm landfall and lifesaving decisions by marrying earth‑observation AI with high‑resolution aerial baselines and smarter address data: satellite AI that automatically compares pre‑ and post‑storm imagery can flag damage over an entire island and - per recent research - spot patches of damage with about 84% accuracy, producing island‑wide alert polygons for responders (satellite AI disaster mapping for hurricane damage); meanwhile, Puerto Rico teams that used Hexagon's 15‑cm orthorectified aerial imagery after Maria found that sharp pre‑event photos let decisionmakers evaluate damage, identify access routes, and triage infrastructure faster than blunt basemaps (Hexagon 15‑cm orthorectified aerial imagery case study for Puerto Rico disaster response).

Those technical gains must be coupled with cleaner location records - after Irma and Maria, imperfect address data and more than 42,000 landslides hampered relief efforts, spawning a federal working group to improve Puerto Rico address data - so maps become actionable for field crews and local guides, not just pretty overlays (Puerto Rico emergency response address data lessons).

The practical payoff is simple and sharp: precise damage polygons plus reliable addresses mean trucks and rescuers reach the right neighborhood sooner when every hour counts.

“The imagery facilitated a measured response grounded in facts that enabled decision makers to evaluate the extent of damage, gauge potential access routes, and determine the scope of infrastructure damage.”

Drafting an island‑wide smart‑lighting RFP for Puerto Rico is a choreography of clear requirements, enforceable evaluation criteria, and airtight compliance - and modern AI writing assistants like Gemini or Microsoft Copilot can speed that choreography by generating first‑draft scopes, reusable templates, and side‑by‑side evaluation matrices that reviewers can refine.

Ground the RFP in procurement best practices (aiming for "best value" across price, quality, and performance) and the federal‑fund rules that matter on the Island - use the VLCT Model Procurement Policy for government procurement thresholds and vendor outreach as a template for thresholds, micro‑purchase/simplified acquisition/large‑purchase rules, conflict‑of‑interest language, and small/socioeconomic vendor outreach.

Pair AI drafting with proven process: adopt RFP templates, set precise SOW and scoring rubrics, and run pre‑solicitation Q&As to avoid ambiguous requirements (How to Simplify the Government RFP Process - PrometSource).

Crucially, keep a human in the loop to verify compliance (2 CFR/agency rules, geographic‑preference permissions) and bake in FedRAMP/ISO hosting and data‑security checks before publishing so the winning smart‑lighting solution is both competitive and deployable on day one (FedRAMP and ISO hosting and data‑security compliance guidance).

The payoff is concrete: a focused, auditable RFP that turns vendor confusion into clear bids and gets lamps on the right streets faster.

San Juan's Government Transparency Office can use Microsoft 365 Copilot to make hiring faster, fairer, and more transparent - generating standardized, bilingual job descriptions, producing interview question sets tied to competency frameworks, and summarizing applicant responses so HR staff spend less time re‑reading and more time verifying fit; this matters in Puerto Rico where many public roles explicitly seek Spanish‑English fluency, from bilingual customer service positions to specialized remote roles listed for the Island (see recent bilingual customer service job listings in Puerto Rico on recent bilingual customer service job listings in Puerto Rico on VirtualVocations).

Copilot‑enabled templates also help auditors trace hiring decisions and redline public postings for compliance, a useful guardrail when agencies must align procurement and staffing with territory rules and cloud security expectations - see why FedRAMP and ISO security compliance matter for government AI adoption.

The result: clearer candidate pipelines, faster bilingual outreach, and hiring records that stand up to public scrutiny while keeping human reviewers in control.

“Working for HSI is about making a difference - not only for our nation, but for our island. As a Puerto Rican woman, I am honored to have selected this career at a young age and to have directed all my energies into contributing to the security of my nation. I encourage you to join our team and develop to your fullest potential as part of your country's security team.”

Monitoring telemarketing outreach against the FTC's Do Not Call registry should be reframed in Puerto Rico as a routine compliance audit - one that pairs suppression‑list checks with strong data governance and transparent procurement practices so outreach vendors don't inadvertently trigger island‑specific scrutiny; the recent civil‑society review of Puerto Rico's governance highlights persistent transparency and enforcement gaps that make proactive audits and clear audit trails essential (UNCAC parallel report: Puerto Rico transparency and enforcement gaps).

Treating suppression checks as an auditable workflow also mirrors the broader enforcement climate on the Island - illustrated by heightened IRS investigations into Act 20/22 decree holders - which shows regulators will follow patterns of noncompliance across domains (IRS audits and investigations of Puerto Rico Act 20 and Act 22 decree holders).

Finally, bake FedRAMP/ISO‑grade hosting, retention and role‑based access into any telemarketing audit plan so suppression lists and call logs remain trustworthy evidence rather than scattered spreadsheets (FedRAMP and ISO security compliance for telemarketing audit plans) - because in practice a single missed suppression can turn routine outreach into a high‑stakes review.

Building municipal staff capacity in Puerto Rico means more than one-off demos: combine hands-on tool practice with accessible, public‑sector courses and bilingual design to make AI useful on day one.

Free, self‑paced training from InnovateUS - courses like "Using Generative AI in Government" and workshops such as “Beyond Translation: Human‑Centered Language Access with AI” - offers practical lessons, risk checklists, and prompt examples that help clerks and field staff safely use Microsoft Copilot or Gemini to produce clear Spanish/English guidance, standardized FAQs, and role‑based playbooks; see the InnovateUS public-sector AI training series and workshops for course and workshop details (InnovateUS public-sector AI training series and workshops).

Pairing that with campus‑scale models of AI fluency - like the Ohio State AI Fluency initiative - creates a repeatable pathway for bilingual knowledge bases, training cohorts, and coach‑led clinics so municipal teams can move from experiments to auditable, language‑aware services without leaving human reviewers behind (Ohio State AI Fluency initiative and resources).

“Through AI Fluency, Ohio State students will become 'bilingual' - fluent in both their major field of study and the application of AI in that area.”

Puerto Rico stands at a decisive moment: new Senate bills drafted in January 2025 propose an AI Officer, an AI Advisory Council, mandatory agency reviews and a registry of AI‑using businesses - concrete levers to make transparency and accountability operational rather than aspirational (Puerto Rico Senate AI Officer and Advisory Council bill (January 2025)).

Coupling that legislative push with proven governance playbooks - AI inventories, risk assessments and NIST/ISO‑style controls recommended in recent state guidance - lets agencies scale pilots into dependable services without leaving equity or security behind (State guidance on AI governance programs and NIST/ISO controls).

Practical steps are clear: establish roles and biennial auditing, bake FedRAMP/ISO‑grade hosting and human‑in‑the‑loop reviews into workflows, and invest in rapid reskilling so systems are used well; Puerto Rico's hiring pilot (which drew thousands of applications in weeks) shows the appetite and urgency for trained staff.

With policy, process, and people aligned, AI can speed services - from disaster response to permitting - while keeping public trust intact.