Show Notes
AI governance for agencies isn’t a policy sat in a folder – it’s what actually happens when your team is using AI and nobody’s watching. In this episode, Gareth puts Callum, our AI lead, in the hot seat on why so many agencies think they’re covered and aren’t: the free AI plan that led to a marketing agency facing a massive insurance claim, the unlocked Google Workspace connector that handed an employee ammunition for an employment tribunal, and why “all or nothing” AI connectors can quietly expose a business’s financials, HR records and boardroom conversations.
In this episode
- What AI governance actually means for an agency, beyond a written policy
- Why “shadow AI” – staff quietly using free or personal AI plans for company work – is still the biggest risk Callum sees, a year into doing this work
- The real story behind a marketing agency’s insurance claim after client data leaked from a personal AI account
- Why having the right AI plan and policy in place doesn’t guarantee anyone’s actually using it (and the audit that proved it)
- How AI connectors to Office 365 or Google Workspace can expose far more than intended – including the case of an employee using leaked meeting data at an employment tribunal
- Why highly regulated agencies (health, finance and similar) should isolate rather than filter, and where to start instead
- Why governance is what lets an agency move faster with AI, not slower
