AI, trust and communications
Your organisation is already using AI, whether or not the board has decided to. Someone on your team pasted something into a chatbot this week. Adoption moves at the speed of a free trial and governance hasn't kept up. If you hold sensitive data and answer to a board, the stakes run higher than most: you run on trust, and there's no technical team to hand the problem to.
None of this needs to be alarming. It needs to be governed, and that's a judgement I make daily, using these tools myself: excited by what they make possible, clear-eyed about what they cost.
Get in touch →Exploring AI tools for your team? Start with The Edit.
An AI tools directory I built and run for charity, cultural and heritage comms teams. Every tool has been through the same checks: where the data sits, whether it trains on your content, and how likely typical use is to trigger a DPIA, the data protection assessment your organisation carries out. The ones that didn't hold up are published too.
AI governance
I work with chief executives, senior teams and boards on adopting AI responsibly and being able to answer for it. That starts with finding where AI is already in use, including the tools nobody approved, because you can't govern what you haven't found. From there: a policy short enough that people actually read it, data matched to the sensitivity of the tool, and a plan for AI-enabled reputational harm decided before you need it, not drafted mid-incident.
Most AI governance is written for organisations with in-house counsel and a technical team. You probably have neither. So I build governance a non-specialist can run: one page rather than forty, a named owner rather than a committee, and a board that can see what's in use and what would happen if it failed.
Training built on your team's real work
Most AI training makes a room feel busy for a day and changes nothing by Friday. Mine is built on your team's actual tasks, gathered through a short diagnostic before anyone walks in, and it carries the honest version: what these tools do brilliantly, what they cost, and where they don't belong.
For leadership teams.
The working understanding leaders need to set direction and govern adoption, without touching a tutorial. A board version closes with a written summary.
For the whole team.
From a hands-on half day to a full day together. Prompts written as professional briefs on real tasks, tested live. The half day configures each person's setup; the full day plans it.
For chief executives, one to one.
A short private series for leaders who would rather build fluency on their own work, confidentially, than in a room with their team.
For the people who are accountable.
Data handling matched to real sensitivity, the difficult scenarios argued out in the room, and your own one-page AI-use policy drafted, with a named owner, before you leave.
Every session ends with something configured, something committed to, and a check-in already in the diary. Sceptics are welcome in every room; most of their objections are right. The exact shape of a session gets worked out at proposal stage, once I've seen the diagnostic.
Where to start
You don't need to know which of these fits before we talk. Working that out together is what the discovery call is for: 30 to 45 minutes, and a rough sense of what the organisation is dealing with is enough. Pricing is scoped per engagement and arrives at proposal stage. If a call feels early, start with the AI-use policy template instead. It's free, it downloads straight away, and it's the document I use as a starting point with clients.