Observations, arguments and pattern-recognition from the inside of UK SME decisions about AI adoption and innovation management. Published first on LinkedIn, kept here in their full form.
The cost of building has collapsed. The cost of ownership hasn't. For every quietly-essential AI tool inside your business, someone needs to own what happens when it breaks.
Read this perspective→
An MD with a live AI tool inside the core sales workflow couldn't tell me what data it was touching or where it went. The architecture sat with one employee. Accountability sat nowhere.
Read this perspective→
The UK government's new Cyber Resilience Pledge isn't trying to regulate every SME directly. It's pushing baseline cyber standards through enterprise procurement. The pressure point has moved.
Read this perspective→
Three online tools, three radar charts, three nudges towards adoption. None of them allowed for the answer 'don't do this yet' — which is the answer SMEs need most.
Read this perspective→
Last week an AI coding agent wiped a production database in ten seconds. The technology didn't fail. The scope decision was never made.
Read this perspective→
Microsoft has named its price. Salesforce, AWS and OpenAI have named different ones. The build-versus-buy decision is no longer just about today's bill.
Read this perspective→
Governance does not simply oversee innovation activity, it shapes it. Apply decision structures designed for certainty to work defined by uncertainty, and the work itself changes.
Read this perspective→
Innovation is increasingly limited not by ideas or technology, but by how leadership teams govern uncertainty, allocate responsibility and make decisions over time.
Read this perspective→