AI Governance Should Help Organizations Move Forward, Not Hold Them Back

Shawn Petti
November 9, 2026

Reflections from the Roc Leads The Shift AI Summit

By Shawn Petti, President

Last week, I had the opportunity to attend the Roc Leads The Shift AI Summit and serve as a facilitator for a breakout discussion focused on AI governance.

Our table included business owners, HR professionals, college professors, technology leaders, and others, all thinking about the same fundamental question:

How do we embrace the opportunities AI creates while making sure we are using it responsibly and safely?

There wasn’t one universal answer, and that was probably the most valuable part of the conversation.

AI Adoption Is Already Happening

One of the biggest things organizations need to recognize is this:

Your employees may already be using AI, whether your organization has an AI strategy or not.

People are using AI to write emails, summarize documents, brainstorm ideas, analyze information, create content, and solve problems.

That creates tremendous opportunities, but it also raises important questions:

  • Which AI tools are approved?
  • What information can employees safely enter?
  • What happens to that data?
  • How do we protect confidential, customer, employee, or regulated information?
  • Are employees trained to verify AI-generated information?

Our discussion reinforced that governance doesn’t have to mean stopping people from using AI. Good governance can actually give employees more confidence to use it because they understand the boundaries.

Governance Should Be a Guardrail, Not a Roadblock

One of my favorite ideas from the discussion was thinking about governance as more of a gas pedal than a brake pedal.

Organizations certainly need boundaries. But if the rules are too restrictive or confusing, employees may either avoid AI entirely or use tools outside of the organization’s visibility.

A better approach is to make expectations clear:

These are the tools we trust. This is the information you can use with them. These are the things you shouldn’t share. And here is how we want you to experiment.

It’s a little like getting a driver’s license. Give people training, establish the rules of the road, provide appropriate tools, and then allow them to safely move forward.

Protecting Data Has to Be Part of the Conversation

Data privacy and security were understandably major parts of our discussion.

Someone copying information into an AI tool may not realize they are potentially moving organizational data into another system. Even a seemingly harmless spreadsheet could contain employee, customer, financial, health, or other sensitive information.

That’s why organizations need to understand the difference between employees experimenting with public consumer AI tools and using properly configured business or enterprise AI platforms with appropriate protections.

Technology alone isn’t enough, though. People need to understand what information is appropriate to share and why.

Don’t Forget the Human Side of AI

Some of the most interesting parts of our discussion weren’t technical at all.

We talked about deskilling, where people become so dependent on technology that foundational skills begin to weaken. We also discussed AI fatigue.

If businesses suddenly mandate AI for everything, introduce new platforms constantly, or use AI simply to create more activity, employees can quickly become overwhelmed.

The goal shouldn’t be to make people busier.

AI should remove low-value work so people have more time for problem solving, collaboration, higher-value work, and human interaction.

You Don’t Need a 50-Page AI Policy to Get Started

For small and midsized organizations especially, AI governance can sound intimidating.

It doesn’t have to be.

Our conversation highlighted a simple framework:

1. Learn
Understand how AI is already being used inside your organization. Don’t assume the answer is, “We aren’t using it.”

2. Translate
Turn security, privacy, and compliance requirements into guidance employees can actually understand.

3. Connect
Bring together the right perspectives, including leadership, HR, IT, legal, security, compliance, and the employees actually using AI.

4. Act
Establish approved AI tools, basic rules for sensitive information, employee training, and a few well-defined use cases where AI can provide real value.

You can mature your approach as your organization’s use of AI grows.

The Goal Isn’t Perfect AI Governance

One of my biggest takeaways from the summit is that organizations shouldn’t wait until they have everything figured out before exploring AI.

At the same time, allowing everyone to choose their own tools and practices creates unnecessary risk.

There is a middle ground.

Start experimenting. Put reasonable guardrails in place. Train your people. Protect your data. Learn what works. Then adjust as you go.

Good AI governance shouldn’t make employees afraid to use AI. It should give them the confidence to use it responsibly.

Instead of only asking:

“How do we control AI?”

Maybe the better question is:

“How do we create an environment where our people can safely use AI to do better work?”

That was the conversation I appreciated most at Roc Leads The Shift, and it’s one I expect more organizations will be having very soon.

About the Author

S

Shawn Petti

Expertise in cybersecurity and helps businesses implement robust security strategies.