MD of Bighorn considers Copilot amid push for AI policy and training

MD of Bighorn is updating its comms policy and piloting Microsoft Copilot across 365 for drafting, summaries, and data work. Governance, training, and clear guardrails come first.

Categorized in: AI News PR and Communications
Published on: Oct 23, 2025
MD of Bighorn considers Copilot amid push for AI policy and training

MD of Bighorn explores internal AI use-what PR and communications teams can learn

The MD of Bighorn is updating its communications and social media policy and considering AI for internal use. The focus: using Microsoft Copilot to help draft documents, analyze data, summarize public information, and improve productivity across Microsoft 365.

"The current policy was written in 2015 and that was the last time that it was amended. Since then, there have been significant changes in social media communications and the development of AI," said MD CAO Shaina Kelly. "We felt that that was also a component that was necessary to incorporate into a policy for the municipality so that we can have some governance pertaining to the use of AI when it comes to communications."

What AI is on the table

Microsoft Copilot is the proposed starting point due to enterprise-grade data protections and its fit with the MD's Microsoft 365 environment. According to administration, Copilot would not learn from MD data and would maintain enterprise protections across the suite. For context, see Microsoft's guidance on privacy for Copilot for Microsoft 365 here.

Governance and compliance come first

In the draft reviewed in August, all AI use must comply with applicable laws, regulations, and internal data governance policies. While no AI governance policy currently exists, the MD is updating internal data policies to include AI.

"Down the road, we will need to look at developing an AI policy if we are going to start utilizing AI for operational purposes," said Kelly. Any future use must comply with laws such as the Access to Information and the Protection of Privacy Act, and the policy will need flexibility to adapt to future provincial or federal changes, including references to items such as the now-terminated Artificial Intelligence and Data Act.

For comparison, public bodies in Alberta often reference the Freedom of Information and Protection of Privacy (FOIP) framework. See the provincial overview here.

Training and detection are on the agenda

Coun. Jen Smith raised two practical questions: who will oversee the AI portion of the policy, and who will be trained to detect and verify proper AI use in policy creation. "This is technology that is very useful when the proper training has been had… It can be very undetectable unless you are trained," she said.

Administration noted that AI detection tools exist, but more research is needed before recommending specific solutions. "Basically, we are starting with Copilot because of the enterprise data protection," said Thomas Judek, adding that Copilot is the only approved use case at this stage.

Process before rollout

Reeve Lisa Rosvold supported creating a corporate AI policy before adding AI into the communications policy. She suggested bringing both forward together so council can see how they connect in practice.

Judek added it may be best practice to update the policy annually to keep pace with industry changes. The draft was accepted as information and will return for further deliberation.

Action checklist for PR and communications leaders

  • Define approved AI tools, use cases, and clear "no-go" zones (e.g., confidential data, legal advice).
  • Set a human-in-the-loop review for public-facing drafts and external communications.
  • Create a disclosure rule for when AI assists with drafting or summarization.
  • Map data classifications and permissions inside Microsoft 365 before enabling Copilot features.
  • Stand up training for policy owners and content editors on prompt quality, bias, and fact-checking.
  • Pilot detection and QA workflows where policy or legal content is involved; document exceptions.
  • Run privacy impact assessments for new AI uses; log decisions and model configurations.
  • Publish an AI governance playbook with RACI roles, incident handling, and an annual review cycle.
  • Track metrics: time saved, error rates, rework, compliance exceptions, and stakeholder feedback.

If your team needs structured upskilling on AI for communications, explore role-based learning paths at Complete AI Training.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)