Hi everyone,

Chad Barendse here, co-founder of DGX, and welcome to the May newsletter.

ADG 2026 is now in full swing. Planning has stepped up a gear and I can already tell you this year is going to be bigger and better than last. We have some genuinely amazing speakers locked in, people doing the real work of governing AI inside complex organisations, not just talking about it. I will share more of the lineup in the coming newsletters, so keep an eye out.

But the thing I have been thinking about most this month is a phrase that gets used far too loosely.

Human-centred governance.

Most of the time it means very little. For me it comes down to one shift, where you start. Traditional governance starts with the framework. You design the operating model, appoint the stewards, publish the policy library, then ask the business to comply. That makes governance feel like something that happens to people rather than something that helps them. It becomes the team that says no.

Human-centred governance starts somewhere else. It starts with where data actually hurts in someone’s day, or impacts customers.

When I work with a team I do not open with a framework. I ask what is slowing them down, what they are worried about, what challenges they are facing. I have seen the difference this makes firsthand. The teams I helped solve a real problem came back the next time with the next one. The teams I handed a policy either didn’t care or found ways to work around it. Same controls, completely different outcome.

To be clear, human-centred does not mean permissive. The lineage, the quality standards, the ownership, none of that disappears. What changes is how you earn the right to enforce it. You earn it by proving you understand the problem first.

If you are thinking about where to start, three things:

• Open every governance conversation with the problem, not the policy. Understand what people are trying to do before you tell them how to do it.
• Measure adoption, not coverage. A policy nobody follows is not governance, it is paperwork.
• Point to named problems solved. Not maturity scores, not policy registers. Real problems, real people, real outcomes.

I went deeper on what human-centred governance actually means in a piece this month. If this resonates, it is worth a read.

Read the full post: