Yes, insurance agencies can use AI. The main insurance AI guidance doesn't ban it; it sets expectations for how it's governed. That guidance, the NAIC Model Bulletin on the Use of Artificial Intelligence Systems by Insurers, is addressed to insurance companies, not agencies. It reaches agencies indirectly, though: carriers are expected to oversee the third parties and AI tools involved in their business, and the bulletin covers the whole insurance life cycle, including sales and distribution. The practical takeaway for an agency is to know which AI tools you use, keep people responsible for the decisions that matter, and be ready to explain your setup when a carrier asks.
General information, not legal advice. Your state insurance department and your attorney can speak to your specific situation.
What is the NAIC AI bulletin?
The National Association of Insurance Commissioners (NAIC) adopted the Model Bulletin in December 2023. It is a template that each state's insurance department can adopt. As of April 1, 2026, 24 states and the District of Columbia had adopted it, according to the NAIC's own adoption map.
The bulletin sets expectations for how insurers govern AI. It defines an AI system as a machine-based system that can generate outputs such as predictions, recommendations or content that influence decisions. Its main concern is "adverse consumer outcomes": AI-driven decisions that break existing insurance laws, such as rules against unfair discrimination.
Does the NAIC AI bulletin apply to insurance agencies?
Not directly. The bulletin is addressed to "all insurers" licensed in the state, and it doesn't mention agents or producers. Three things still make it relevant to agencies:
- It covers distribution. The bulletin says AI is used across the insurance life cycle, including marketing, sales and distribution, and it asks insurers to govern AI across that whole life cycle.
- It expects insurers to oversee third parties. Insurers are asked to do due diligence on third-party AI systems and data, to include audit rights and cooperation with regulators in their contracts, and to monitor compliance over time.
- Existing laws still apply to everyone. AI doesn't change the rules on licensing, unfair trade practices, privacy or discrimination. If a task needs a licensed producer today, it still does when AI helps with it.
The good news: none of this stops an agency from using AI for everyday operations. It mostly asks for the same care you'd already take with any system that touches client information or advice.
Do some states have their own AI rules for insurance?
Yes. A few states have gone further than the model bulletin. New York's Department of Financial Services issued Insurance Circular Letter No. 7 on July 11, 2024, covering insurers' use of AI and external consumer data in underwriting and pricing. Colorado has its own law, SB21-169, aimed at unfair discrimination from insurers' use of external consumer data and algorithms. Both are directed at insurers, but they show where regulators are focused. Check your own state insurance department's website for current guidance.
What AI uses are lower risk for an agency?
The bulletin's focus is on decisions that affect consumers: who gets coverage, at what price, and how claims are handled. Most agency automation sits far from those decisions:
- Sorting inbound emails and service requests
- Sending renewal reminders and collecting missing documents
- Updating your agency management system or CRM after a call
- Drafting routine replies for a licensed team member to review
- Preparing certificate of insurance requests for review
Uses that deserve more care are anything that recommends coverage, influences eligibility or pricing, or speaks to clients without a person reviewing it.
What should an agency do to use AI responsibly?
Most agencies can cover the basics in an afternoon:
- List the AI tools you use, including AI features inside software you already have, and what each one does.
- Keep a person responsible for coverage advice, eligibility and anything a license is required for.
- Ask your vendors three questions: where does our data go, who can see it, and can you give us documentation if a carrier or regulator asks?
- Write down your process in a page or two: what the AI does, who reviews it, and how errors get caught.
- Expect carrier questions. As carriers build their AI programs, they may send agencies questionnaires or update their agency agreements. A short written record makes those easy to answer.
None of this requires a compliance department. It's the same good record-keeping that makes any system easier to trust and improve.
How does White Feather AI approach this for agencies?
We start with one operational workflow, like service requests or renewals, that doesn't make coverage or pricing decisions. Your team sees and approves every step, we document what the system does, and you own it after handoff. See our services or read what an AI agent is. For questions about data security, see our FAQ.
Wondering if your agency is ready? Take the 2-minute readiness quiz.
Frequently asked questions
Is it legal for an insurance agency to use AI?
Yes. Neither the NAIC bulletin nor the New York and Colorado rules ban agencies from using AI. Existing insurance laws, like licensing and unfair discrimination rules, still apply to anything the AI helps with.
Has my state adopted the NAIC AI bulletin?
As of April 1, 2026, 24 states and DC had adopted it. The NAIC publishes a current adoption map, and your state insurance department's website will list its bulletin if it has one.
Does the NAIC AI bulletin require agencies to have an AI program?
The bulletin asks insurers, not agencies, to maintain a written AI program. Agencies may still be asked by their carriers to describe the AI tools they use.
Can AI give coverage advice to clients?
That's where to be most careful. Keep a licensed person responsible for coverage recommendations, and use AI to prepare information for them rather than to advise clients on its own.
What is an adverse consumer outcome?
In the NAIC bulletin, it's a decision by an insurer that adversely affects a consumer in a way that violates insurance regulatory standards, such as unfair discrimination.