AI Policy & Governance

Why this stopped being optional

Three things changed. Clients started asking vendors how they use AI on their work — and putting it in contracts. Insurers started asking. And staff started using AI tools with company and client data whether or not anyone approved it.

When the Greater Sarasota Chamber of Commerce ran its first AI summit for member businesses, the panel led with legal ramifications, copyright and misinformation. That is the tenor of the conversation locally: interest paired with real, well-founded caution. A written policy is how you convert caution into permission to move.

What we produce

• Acceptable use policy — approved tools, prohibited tools, and what may be entered into each, in plain language
• Data classification — which categories of information may go where, with your actual data types named
• Human review requirements — which outputs need sign-off before they leave the building, and by whom
• Client disclosure standards — when and how you tell clients AI was involved in their work
• Vendor assessment checklist — the questions to ask any AI vendor before signing, including the ones they hope you skip
• Incident response — what to do when something goes wrong, because eventually something will
• Review cadence — a schedule for revisiting the policy as tools and rules change

Sector-specific requirements