Ai at work · 8 October 2026 · 3 min
You Were Told to Use AI. Nobody Told You the Rules.
The question
Why do so many professionals use AI at work without clear guidance from their employer?
The short answer
Because adoption moved faster than guidance. Write down your own working rules for one task this week and take them to your manager.
You use AI at work. You probably found the tool yourself. And nobody has told you what you may and may not do with it.
You are not the exception. You are the pattern.
What the research shows
Deloitte research published in September 2026, based on a survey of 25,000 UK workers carried out by Ipsos, found that 63% of working adults knowingly use generative AI for work. Nearly a third of those users say they use it without their employer's knowledge. Around half say they have received no formal training on using AI safely and effectively. And 65% report a lack of convincing leadership on how it should be used.
Germany shows the same shape. The third wave of the Konstanz AI study by the University of Konstanz found that only 55% of AI users say their most used tool was officially introduced by their employer. In small organisations, only 11% reported AI training and only 10% reported binding rules.
Small business owners tell a similar story. In a September 2026 survey by Echelon Insights for Heartland Forward, 83% of owners who use AI say they taught themselves by trying it. Only 12% learned through a class or training programme.
Why this matters for you
Three things are true at the same time.
The tool is in your hands. The rules are not on paper. The judgement is still yours to defend.
When something goes wrong, nobody will ask which tool you used. They will ask why you signed it off.
One step for this week
Pick one task where you already use AI.
Write three lines. What you let AI do. What you always check yourself. What you never put into it.
Take those three lines to your manager and ask one question. Is this how we want it done?
That is not a policy. It is a starting point. And it puts your judgement on the record before someone else defines it for you.
What I am learning
Here is what I am learning from this research. The gap is rarely skill. It is clarity. People are already experimenting. What they lack is a shared line between helpful and risky.
You do not need to stop using AI. You do not need to wait for a policy. You need to make your way of working visible.
Write your three lines. Share them. Keep your judgement in the loop.
If you want to see which level of AI judgement your role now asks for, start with the Gap Check at The Cohort.
Which task would you write your three lines for first?
I use AI as a thinking and editing partner. The experience and opinions are mine.