Signal Mountain Town Manager Uses AI for Policy Drafting, Hiring

Signal Mountain Town Manager Matt Justice has integrated artificial intelligence into his daily municipal workflow, using ChatGPT to draft ordinances, resolutions, and job descriptions. The Chattanooga Times Free Press obtained over 500 pages of Justice's AI exchanges through an open records request during Sunshine Week, revealing the extent of his reliance on the technology for basic administrative tasks.

Justice doesn't hide his AI usage, but he doesn't actively disclose it either. "It's here, and it's being used," Justice said by phone. "The newness is worn off of it. I think that folks expect you to use it, especially for the basic tasks." This casual acceptance marks a shift from the initial AI hype cycle to something more mundane—like using a spellchecker, only for entire documents.

The town manager's workflow demonstrates a trust-but-verify approach. When Justice asked ChatGPT to create resolutions terminating a lease with the Signal Mountain Elementary Preservation Fund, the AI produced drafts that closely matched the final versions passed by the Town Council. Some edits were made—numbered lists became body text, for instance—but the bulk of the wording remained identical.

There's a physical reality to this process that matters. Justice sits at his computer, types a prompt, waits for the response to load (a few seconds of screen flicker), then reviews the output line by line. He has to catch errors like when ChatGPT mistakenly called the Mountain Arts Community Center the "Municipal Activities and Community Center." Justice corrected it with a follow-up prompt: "It is the Mountain Arts Community Center. You should know this." The AI apologized and corrected itself.

Mayor Elizabeth Baker supports Justice's AI adoption. "We read the resolutions and the ordinances that he has created, whether he has created them sitting at his computer using Microsoft Word or using AI," Baker said. "He's got to start somewhere." Her perspective reflects a pragmatic view of municipal work—whether it takes 12 hours or 18 hours a day, the work gets done, and saving time is acceptable.

Justice views AI as providing "good bones or good foundation, but it's not complete." He doesn't trust it to deliver finished products. The technology is only as good as the prompt and the information fed into it. This creates a workflow where the human does the heavy lifting of verification, not the initial drafting (which is the real time-saver, though it also creates new risks).

The town lacks formal AI policies for employees. Through a separate open records request, the Times Free Press sought documentation on AI usage guidelines. The town responded that it "does not currently have any written policies, guidelines, directives, memoranda, standard operating procedures, internal guidance, training materials or other records governing or addressing employee use of artificial intelligence, including generative AI, machine learning systems or automated decision-making tools."

This policy vacuum raises questions about consistency and accountability. When Justice had ChatGPT draft an ordinance allowing the town to charge fees for community center use, the final version placed fee-amendment authority under the town manager instead of the Town Council—a significant change from the AI's original draft. The Town Council passed the ordinance in January without Justice disclosing his AI use during the initial discussion.

Regional context matters here. The Times Free Press made dozens of document requests from state agencies and local governments in the Tennessee Valley to examine AI policies and public records compliance. This investigation is part of a broader pattern of local governments adopting AI without clear guardrails.

Justice's approach reflects a broader trend in government AI adoption. According to reporting from NewsChannel9, roughly 500,000 jobs in Tennessee are expected to be augmented or transformed by AI in coming years. The technology is spreading faster than policy frameworks can develop.

The core tension remains: AI provides efficiency gains, but the lack of disclosure and policy creates transparency gaps. Justice doesn't actively hide his AI use, but he doesn't proactively disclose it either. This middle ground works for now, but it leaves residents and council members in the dark about how much of their government's output comes from algorithms versus human judgment.

Whether this informal approach holds up under scrutiny remains uncertain. The town manager's productivity gains are real, but the absence of written policies means every employee operates in a gray area. Someone will eventually make a mistake that AI helped create, and the town will have to decide whether to blame the tool or the person who used it.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn