Grand Traverse County Advances Cloud Migration and Copilot AI for Justice Center Planning

Grand Traverse County is poised to approve a sweeping set of technology investments that will accelerate cloud migration across multiple departments, add 100 Microsoft Copilot AI seats, and launch a formal planning process for a new consolidated justice center — moves that collectively mark a decisive shift toward cloud‑native operations and early enterprise AI adoption for the county government.

Background​

Grand Traverse County’s board of commissioners met to consider several interlocking items: an upgrade of the county’s Enterprise Permitting and Licensing (EPL) system to a cloud model, renewal and expansion of Microsoft 365 subscriptions to include Copilot AI licenses, and authorization of a request‑for‑proposals (RFP) for a consultant to plan a new justice center that would consolidate jail and law‑enforcement functions. These agenda items come after a disruptive ransomware incident that forced the county and City of Traverse City to take networks offline in June 2024 and prompted an earlier migration of the county’s public safety CAD to a cloud environment.
The IT memo prepared by the county’s IT leadership frames the cloud shift as both a security and operational modernization imperative: cloud hosting provides near‑continuous security patching, automated upgrades and geographic redundancy for critical data that on‑premises servers cannot easily match. At the same time, the county is budgeting material recurring costs to maintain that posture and is explicitly investing in staff training and governance for the first phase of AI adoption.

---

Overview of the planned technology moves​

Enterprise Permitting & Licensing (EPL) cloud upgrade​

• What’s proposed: Move the county’s EPL suite (used by Construction Codes, Environmental Health, and GIS/Equalization) from on‑premises hosting to a cloud‑hosted model provided by the software vendor. Tyler Technologies’ Enterprise Permitting & Licensing is the product in question; Tyler positions the product as a cloud‑capable, modular permitting platform for local governments.
• Estimated cost: The county packet lists a first‑year cost of roughly $274,603 with an annual recurring cost of $256,003 thereafter; the county expects annual price growth in the 2–5% range. That compares to the county’s current EPL subscription — slightly over $82,000 annually — meaning the first year’s budget impact is substantially higher.
• Implementation timing: The county expects additional major systems (BS&A and Circuit Court JIS) to migrate to the cloud in 2026, following the EPL upgrade.

Microsoft 365 renewals and Copilot​

• What’s proposed: Approve nearly $400,000 in Microsoft 365 subscription renewals that include 100 new Microsoft 365 Copilot licenses. The county’s IT leadership intends those Copilot seats to establish a set of trained users — a pilot cohort — across nearly 30 county offices and to underpin a “Center of Excellence in Artificial Intelligence” concept.
• Vendor pricing context: Microsoft publicly lists Microsoft 365 Copilot for enterprise at $30 per user per month (annual commitment), which matches widely reported enterprise pricing and helps explain why adding 100 seats is a notable recurring expense. The county’s memo acknowledges the need for training, governance, and accuracy verification as prerequisites to productive Copilot use.

Public safety and AI pilots​

• Prior action: In response to a June 2024 ransomware incident that disrupted county systems, Grand Traverse migrated its public safety CAD to cloud hosting and is experimenting with AI to triage non‑emergency dispatch calls. The county has already contracted an AI call‑handling pilot for non‑emergency Central Dispatch traffic; the goal is to free dispatchers to focus on emergent incidents.
• Operational controls: County leadership states Copilot and other AI pilots will be accompanied by staff training and a forthcoming AI policy to govern permitted use, ethical safeguards, and verification practices.

Facilities planning RFP for a new justice center​

• Scope: Approve an RFP to engage a consultant for pre‑design, feasibility, and (optionally) construction oversight for a proposed justice center that could include a new jail and consolidated law‑enforcement space. The consultant will evaluate options across multiple county campuses and present budget estimates and timelines.
• Phasing: The RFP includes three phases — (1) feasibility/pre‑design and budget, (2) design coordination and permitting, and (3) optional construction oversight pending future approval.
• Fiscal posture: Financial advisors told the commissioners that debt scenarios for the justice center and other capital projects look “manageable” when spread over decades, but the county will still be taking on long‑term obligations that require careful cashflow management.

Treasurer’s audit discussion​

• Context: County Treasurer Jamie Callahan is scheduled to present after the board requested his appearance to discuss the county’s most recent audit. The audit was clean but late; it flagged late bank reconciliations and recommended the treasurer provide regular investment reports. The treasurer’s packet included monthly interest income charts for 2025 through August.

---

Why the county is moving now: context and drivers​

The county’s decisions are shaped by three systemic drivers:

• Security urgency after a ransomware incident: The June 12, 2024 ransomware event forced the county to take networks offline and created operational disruption across courts, payments and dispatch. That incident elevated the operational risk of on‑premises hosting for mission‑critical services and accelerated decisions to adopt vendor‑hosted, continuously patched cloud services for public safety.
• Vendor evolution away from on‑premises support: County leadership reports that vendors themselves are increasingly de‑prioritizing on‑premises product versions in favor of cloud offerings, making cloud migration not just desirable but often necessary to remain on supported product versions.
• Strategic AI pilots and workforce augmentation: Like many municipal IT organizations, Grand Traverse is piloting AI to capture productivity wins in administrative workflows and to integrate AI where it safely reduces staff load — for example, non‑emergency call handling and administrative drafting inside Microsoft 365. Adding Copilot seats and planning a Center of Excellence are early moves to institutionalize governance, measurement and staff training.

---

Technical verification of key claims​

To ensure the county’s technical assumptions are grounded in vendor capabilities and market reality, several claims were verified against independent public documentation:

• Microsoft 365 Copilot pricing: Microsoft lists Microsoft 365 Copilot for enterprise at $30 per user per month (annual commitment), which aligns with industry reporting and explains the county’s budgeting rationale for 100 seats. This pricing is subscription‑based and recurrent — making licensing cost modeling essential for any broad rollout.
• Tyler Technologies’ EPL product: Tyler publicly markets Enterprise Permitting & Licensing as a cloud‑capable, modular permitting suite used by hundreds of jurisdictions; Tyler also promotes cloud features such as GIS integration, online submittals and mobile inspection apps. Migrating to Tyler’s cloud model is therefore technically plausible and consistent with Tyler’s product roadmap.
• Ransomware and cloud migration precedent: Grand Traverse’s June 2024 ransomware incident and the subsequent decision to migrate the county’s CAD to a cloud host are documented in local government communications and in coverage by technology news outlets, which describe the CAD restoration and the reasoning behind moving public safety to a hosted vendor environment. That sequence of events is the proximate cause for the county’s current cloud push.

Where the county supplied specific financial numbers (for example, the exact first‑year cost for EPL), those numbers are drawn from the county’s meeting packet and should be treated as the county’s official estimates; vendors’ final contract proposals and negotiated terms may change those figures.

---

Strengths of the plan — what the county gains​

• Improved security posture for mission‑critical services: Cloud hosting provides continuous patching, managed security services and multi‑region redundancy that materially reduce single‑site failure risk compared with isolated on‑premises servers. After a ransomware event, that resilience is a practical risk mitigant.
• Reduced operational burden on local IT staff: Vendor‑hosted applications shift routine patching, updates and availability SLAs to the vendor, freeing county staff to focus on integrations, policy and user support rather than nightly backups and emergency restore operations.
• Measurable AI pilots with governance built in: Limiting Copilot rollout to an initial cohort and creating a Center of Excellence are best practices for early AI adoption, enabling the county to instrument outcomes, maintain human‑in‑the‑loop controls, and codify acceptable use cases before broad deployment. Microsoft’s Copilot tooling also provides admin controls and audit telemetry to support governance and measurement.
• Consolidated planning for facilities: Bringing in an external consultant via an RFP to do feasibility and cost modeling for the justice center is a prudent step to establish objective estimates, phasing options and a funding strategy before the county commits to construction debt.

---

Risks and open questions — what commissioners should scrutinize​

While the plan has clear merits, the county is taking on recurring costs and governance responsibilities that create new risk vectors.

• Recurring license and subscription costs can scale quickly. Microsoft Copilot is a recurring expense (public pricing shows $30/user/month for enterprise Copilot). A 100‑seat pilot may be affordable, but broad adoption across hundreds of users would multiply that cost substantially. Municipal IT budgets must model not just pilot costs but sustainable long‑term licensing and potential tiered entitlements.
• Contract details matter for security and continuity. Moving to cloud hosting shifts responsibility but also requires robust vendor SLAs, data residency guarantees, and clear incident response terms. Counties should confirm:
• Who controls encryption keys and backups
• CPS/PCI/CJIS/HIPAA compliance obligations (as applicable)
• Egress and portability terms if the county later decides to change vendors or repatriate data
• Pricing escalation caps and termination clauses to avoid surprise cost growth
  These operational contract details are the levers that determine whether cloud migration reduces or merely redistributes risk.
• Vendor lock‑in and integration costs are often underestimated. Even when a vendor hosts the application, integrations to local systems (GIS, tax, courts) frequently require middleware, connector maintenance, and custom development. The county should budget lifecycle integration and testing costs, not just initial subscription fees.
• AI output accuracy and public trust: Generative AI assistants can hallucinate or provide inaccurate summaries if prompts and data connectors aren’t tightly controlled. For public‑facing or legally sensitive tasks (for example, drafting official correspondence, open records redactions, or dispatch triage), the county must maintain human review, logging, and clear standards for acceptable AI use. The planned AI policy and training are essential; implementation rigor will determine whether pilots earn public trust.
• Fiscal risk from capital projects: The justice center planning proceeds alongside several other capital priorities. Although advisors projected manageable debt when spread across decades, construction debt still requires conservative contingency planning and clear prioritization if operating revenues are constrained. Voters and oversight bodies often scrutinize such long‑range borrowing; the county should prepare transparent scenario analyses.

---

Practical recommendations — governance, procurement and implementation checklists​

To translate the agenda items into durable, low‑risk outcomes, the county should require the following before full contract approvals or large rollouts:

• Contract and security checklist for cloud services
• Require explicit data portability and egress pricing language.
• Insist on role‑based access logs, per‑request audit trails and vendor support for tenant‑level encryption key management or escrow.
• Validate CJIS, FedRAMP, HIPAA or other controls as required by the workload.
• Pilot metrics and evaluation gates for AI
• Define SMART KPIs (e.g., percentage of non‑emergency calls handled, percent reduction in dispatcher handle time, accuracy thresholds).
• Require human‑in‑the‑loop verification for any AI output that could affect legal status, public safety or fiscal decisions.
• Publish anonymized pilot metrics periodically for transparency and community confidence.
• Financial modeling and contingency planning
• Model best‑, base‑, and worst‑case licensing trajectories for Copilot and other subscription services over a five‑year horizon.
• Establish a technology reserve or recurring line item to smooth subscription renewals and unexpected price escalations.
• Procurement design for justice center planning
• Require the consultant RFP to include alternative scenarios (renovation vs. new build), life‑cycle operating cost estimates, and a clear public engagement plan.
• Include independent cost‑estimation and value‑engineering tasks to limit scope creep during design phases.
• Workforce and training
• Fund role‑based AI training and change management for administrative staff, including prompt libraries, accuracy verification checklists, and adoption KPIs.
• Align AI pilots with a personnel plan that clarifies whether AI augments roles, reassigns responsibilities, or opens opportunities for reskilling.

---

What the county should measure to demonstrate value​

• Time‑savings per task (e.g., permit intake processing time, dispatch triage handling time)
• Cost per processed item before and after cloud migration (including total cost of ownership: subscription + integration + staff)
• Incident response and mean time to recovery (MTTR) improvements attributable to cloud redundancy
• AI accuracy and error rates, with an incident log for any erroneous outputs that required human remediation
• Adoption metrics (active Copilot users, daily interactions) and qualitative feedback from staff

These measurable outcomes will let the county convert pilot activity into defensible decisions about whether to scale Copilot or other AI tools to more users.

---

Final assessment​

Grand Traverse County’s package of actions — moving permitting to a cloud model, adding a measured set of Copilot licenses, formalizing a justice center planning process, and tightening audit and treasury oversight — is a coherent step toward modernizing municipal operations in the wake of a disruptive cyber incident. The county is following several recognized best practices: prioritizing mission‑critical recoveries, limiting early AI exposure to pilot cohorts, and engaging outside expertise for major capital planning.
Yet the promise of improved security and productivity comes with trade‑offs: recurring subscription costs, contractual and compliance complexity, vendor‑lock risks, and the need for disciplined governance to prevent AI misuse or error. Commissioners should treat the current approvals as program initiation rather than final solutions — approve pilots, harden procurement guardrails, require measurable gates, and fund training and auditability. If the county approaches the next 12–24 months with that discipline, the investments can yield meaningful operational resilience and service improvements for residents.
Grand Traverse County’s move is an instructive case study for other local governments: security incidents accelerate modernization, but successful transformation depends on conservative contracting, measurable pilots, and continual attention to governance and workforce readiness.

---

Conclusion
The commissioners’ agenda reflects a pragmatic pivot — one that accepts higher recurring costs in exchange for resilience, continuous updates, and a pathway to AI‑assisted productivity. What remains decisive is not the technology alone, but the county’s ability to lock good contract terms, to measure outcomes, and to maintain transparent governance so that the public benefits of modernization are demonstrable and sustainable.

---

Source: The Ticker County Eyes AI/Tech Upgrades, Jail Planning Contract