This newsletter provides practical guidance, tools and resources for the real work of governing safe, secure and lawful AI.

Newsletter #56 - April 2026

Feature Article

How ready are you to govern AI?

By James Kavanagh

Certification puts you at the start line, not the finish. It gives you baseline knowledge. What it doesn't give you is the practitioner capability to actually do governance work - the resources to do it well, the influence and judgment that grow from experience. Building that takes sustained effort beyond any certification, and most people underestimate how much.

I keep having two versions of the same conversation. The first is with practitioners who have invested seriously in building their AI governance knowledge. They've built careers in adjacent fields - law, risk, compliance, engineering - and then invested in the IAPP AIGP on top of that. They've done real work. But the role they want isn't arriving, or the work feels harder than expected, and they can't quite see why. The second version is with leaders in organisations who've been told by someone credible that governance means compliance. They're focusing on certification to ISO 42001, or preparing for EU AI Act conformity. Real work, worth doing. And yet when I push a little, it becomes clear that the underlying capacity to actually govern their AI - the visibility, the mechanisms, the culture - hasn't been built.

Neither group has taken a wrong turn. But I think both are further from where they think they are than they realize. And the reason is something a psychologist at Cornell named David Dunning identified decades ago: you need a certain level of capability to accurately assess your own capability. Without it, you can't see the gaps. You reach for something that looks like a solution, it gives you enough confidence to stop looking, and the gap stays exactly where it was.

AI governance isn't a single capability you either have or you don't. It's a set of interlocking capacities that mature at different rates. An organization might have strong leadership commitment but no visibility into its AI systems. A practitioner might have deep technical knowledge but no ability to influence the decisions that matter. Knowing your overall capability needs to rise isn't particularly useful. Knowing that your visibility is strong, your risk sensing is developing, but your integration is almost non-existent? That's something you can act on.

That's what this article, and the assessment we've built around it, are trying to help with.

​Read the full article​

AI Governance Assessment Data

What hundreds of assessments are already telling us about AI governance readiness.

By James Kavanagh

We launched the AI Governance Practitioner Assessment last week. In that time, hundreds of practitioners have already completed at least one - a practitioner assessment and/or an organizational one. It's early, and I want to be careful not to overread it. But the patterns are consistent enough to be worth sharing.

Across the practitioner assessments, Knowledge scores highest. That's not surprising. Knowledge is what most training courses build and certifications test. But Judgment and Influence score noticeably lower. Those are the two dimensions that most determine whether governance work actually changes anything. Judgment is what lets you read a situation and calibrate your response. It allows you to recognise that a generative AI system interacting with vulnerable populations demands fundamentally different governance intensity than an internal analytics dashboard, and to act accordingly. Influence is what lets you walk into a room where legal, product and engineering are pulling in different directions and move people toward a decision. Neither develops through study. Both grow from practice and guidance, from navigating situations where the right answer isn't obvious and the room is waiting for you to say something useful.

The organizational picture is different but related. Across the organizational assessments so far, Capability scores lowest. Visibility, Leadership and Risk score better. That tells me awareness has arrived in a lot of places. Leaders know AI governance matters and have started to act on it. But awareness without the structural conditions to deliver governance is expensive in a specific way: it feels like progress without being it. You can have executive sponsorship, a governance committee and a polished policy document. If the mechanisms aren't embedded in how AI systems are actually built and deployed, the governance exists on paper and nowhere else.

What the data is pointing to, tentatively, is a field that has invested heavily in knowledge and awareness, and is now running into the harder work of building the capability and organisational conditions that make governance actually function. That's not a failure. It's what progress looks like at this stage. But it does mean that the next investment most practitioners and organizations need to make looks very different from the last one.

If you haven't taken the assessment yet, it's free and takes less than fifteen minutes. Results are delivered to your inbox immediately, and the shape of your profile will tell you more than a general score ever could. You can also compare your results to others in your industry and region to understand more.

What we're working on

If you've taken the assessment and want to understand what your results actually mean in practice - how to read the shape of your profile, how your scores compare to others in the field, and what to focus on first - we have two free live webinars running this month. I'll be walking through how to interpret results across each dimension and how to turn what you find into a development plan you can act on. We'd recommend completing the assessment before you join so you come in with your own results to work from. Register for a live webinar here.

We're also building out our Specialty Courses. There are six domain-specific tracks that build on the Foundation Track and go deep in compliance, risk, engineering, evaluation, operations, and leadership. All of them are well underway. AI Compliance will be released first, scheduled for July 2026 release. If you know which specialty is most relevant to your work, join the waitlist now and we'll notify you 24-hours before we open to the public. Your preference will drive our priority. Explore our specialty courses here.

As always, if anything in this newsletter or in any of the work you're doing with us has raised questions or sparked something in you, please reply to this email. We love hearing from you.

PS. You're receiving this as a subscriber to communications from AI Career Pro. We respect your privacy, so please unsubscribe through the link below if you do not wish to receive these communications in the future.