Capability Loop.
Capability Loop is about the real work of AI transformation, not prompts or feature updates. This newsletter and podcast focus on the people doing this day to day, discussing what it takes to build AI capability across an organisation, how workflows get codified, the governance that makes it safe to scale, and the measurements that prove it worked.
Adoption is not capability.
Most organisations have the tools. 88% are still at the early stages of AI adoption, and 55% of AI decision makers say their organisation is investing in AI faster than employees can learn it. That gap is where the investment quietly leaks away.
Capability Loop is an honest look at what it takes to truly build organisation-wide AI capability: written up every issue in the newsletter, and talked through with the people doing it on the podcast.
Source: Notion Global AI Transformation Study (2026), n=6,118 across 10 global markets.
Written for the people who answer for it.
CEOs, MDs, people leaders and CIOs carrying the AI question for their organisation.
You'll likely recognise yourself in this:
The tools are rolled out, usage looks healthy enough on the dashboard, and you still can't say, with evidence, whether your teams are genuinely better off. If you recognise yourself in this, then Capability Loop is for you.
One loop, two formats.
Capability Loop, written.
Four strands an issue, always finishing on the thing you can act on and how you'd know it worked. Free to read, free to subscribe, and one click to leave if it isn't for you.
Capability Loop, in conversation.
Andy in conversation with the people running AI transformation day to day: the workflows they codified, the governance that let them scale it safely, and how they measured whether it worked.
Four strands, every issue.
First-hand, from inside the work.
Andy writing about building AcademyAI and working with the organisations using it: the decisions made with incomplete information, what worked, and what didn't. Honest rather than polished.
What the industry is saying.
What actually happened in AI: what the research is saying and how organisations are transforming operations to become AI-native at scale.
What good actually looks like.
How organisations further along than most are structuring teams, governing AI, and measuring whether any of it is working. Specifics you can borrow, not case-study gloss.
New models, and what they change.
Model releases and significant platform updates, translated into implications: which workflows shift, which risks move, and what it means for your roadmap.
The same loop, every time.
Every issue runs the same four moves, so you always know what you're getting and where to stop if a section isn't yours. News that ends in an observation is entertainment. We'd rather finish on the thing you can act on, and how you'd know it worked.
What actually changed
What it means for you
What to do about it
How you'd know it worked
Read it, or listen to it.
The newsletter is on Substack, the podcast is on Spotify. Free either way, and one click to leave if it isn't for you.