Your AI Policy Is Only Half the Playbook

By Bryce Stuckenschneider | Published 2026-09-28 | Industry | 4 min read

You need the rulebook, and you need a game plan. An AI Charter explains what your organization intends to do with AI and what it’s willing to be held accountable for.

Ask a leadership team what they’re doing about AI, and you’ll often hear about the policy. Which tools can employees use? What data can they put into them? Who reviews the output? What happens when something goes wrong?

Those are important, no doubt. A good AI Policy gives people rules they can use in the middle of a workday, including a safe path to try something new.

But 95% of companies we talk to just end with the "don't do this" AI Policy (which is unfortunately often written by a scared person in the IT department, not a visionary). There's a balance to a good AI policy though... I think great companies need an AI Charter alongside it.

An AI Charter is where the organization says what it intends to do with AI and what it’s willing to be held accountable for. Will we give people time to learn? Where are we comfortable taking a calculated risk? Which decisions or processes will always need a living, breathing person in the middle of them? How will we tell whether a new tool helped our employees and customers, or just made a bad process faster?

Those commitments change what happens when someone on the team has an idea.

The policy might say that sensitive customer information stays out of unapproved tools (very standard language, by the way). The charter can commit the company to providing approved tools so people have a way to experiment safely. The policy might require human review before an AI-generated recommendation reaches a customer. The charter can explain why the company wants a person responsible for that recommendation.

You need the rulebook, and you need a game plan. If the only message from the sideline is “don’t draw a flag,” people are going to be pretty hesitant to go for it on fourth down. And if you're a fan of the Lions or Bears, you LOVE going for it on fourth down.

Give people something they can hold you to

A charter should be specific enough that an employee can point to it later. If the company says it will help people build AI skills, the team should see training and time to practice. If it promises to keep people at the center of the work, leaders should be able to explain how employees and customers are affected when a process changes. If it says humans remain accountable, someone should be able to name the human. I know that may sound pretty obvious, but I'm willing to bet your organization doesn't have a charter (odds I can't get on DraftKings, yet).

I also think prospective employees are paying attention. Someone who already knows how to build with AI wants to know whether they’ll have room to do good work, access to the right tools, and leaders who will make thoughtful decisions. A list of restrictions tells them part of the story. A charter lets them see the company’s ambition and the responsibility that comes with it. If I was considering a new job and I was told "you have basic Copilot, we offer no training, and we review all requests for other tools once per year," I would quickly reconsider.

But if they had a charter that said "we're open minded, willing to try new things, believe in our policies but even more in our people," then I'd get a haircut before the interview.

Put the two documents in the same conversation. Use them when you onboard people, evaluate a new AI use case, and review what’s working. The charter should give the policy a purpose people understand. The policy should make the charter’s commitments real in day-to-day work.

I think companies need both. People deserve to know the boundaries, and they deserve to know what the organization is willing to build together.