A colleague in accounts has a list of two hundred clients in front of her and needs to pull out the ones more than three months overdue. Doing it by hand would eat her afternoon. So she pastes it into ChatGPT and twenty seconds later it's done. She did good work, quickly, and told nobody — it didn't occur to her that she should.
That moment now happens in companies every single day. And it has a name: shadow AI — the use of AI tools a company doesn't know about and doesn't govern. When sensitive data goes into those tools, the company may not even know it has a security problem.
What IBM's data shows
The IBM Cost of a Data Breach Report 2025 studied breaches at 600 organisations. It highlights a gap between the speed of AI adoption and security oversight.
Organisations with high levels of shadow AI had average breach costs $670,000 higher than those with low or no shadow AI. This is an association in the 2025 data, not proof that unapproved AI use alone caused the difference.
In the same study, 63% of surveyed organisations had no AI governance policy in place or were still developing one. That figure cannot be applied automatically to every business, but it is a reason to find out which tools people actually use.
Rules without visibility aren't enough
Rules help only when they reflect what people actually need and use at work.
In practice, I see a policy circulated by e-mail alongside everyday use of other tools. Without speaking to staff and keeping track of the services in use, it is hard to decide which data the company must protect.
The research does not prove that bans fail. It does show why rules should be paired with a usable, approved alternative and a clear view of where company data goes.
Why a ban alone doesn't hold
When a company bans AI and offers no alternative, it puts employees in front of a choice: work slower, or work around the rule. Most people choose the work — especially with a deadline and nobody watching.
I understand the other side of it too. A ban is the fastest option for management because it requires no decision about tools, budget or training. You write "don't use this" and the responsibility shifts to the employee. But the risk doesn't go anywhere — it just stops being visible.
The second problem is that a blanket ban throws away the benefit as well. That colleague has a genuine need AI handles well. A ban doesn't remove it — it just guarantees the solution happens outside company oversight.
What it means legally
This is where it shifts from "security risk" to "legal liability", in two layers.
GDPR. When an employee processes client personal data as part of the company's work, the company generally remains the controller and must put appropriate safeguards in place. Processing solely for the employee's own purposes may be assessed differently; European Data Protection Board guidance notes that an employer may still bear some responsibility if inadequate security enabled it. A ban alone is no substitute for managing access to personal data.
The AI Act. The obligation to support staff AI literacy has applied since February 2025; supervision began in August 2026. A company should therefore know which AI systems are used in its work and help staff understand their limits. I covered the wider framework in my piece on the AI Act in 2026.
The practical point: when AI is used in company work without oversight, first establish what data is processed and who is responsible for the activity. The legal answer depends on the circumstances.
What to do — in order of sequence, not importance
1. First find out what's actually happening. Not through a policy, through a conversation. Ask people what slows them down and whether they're already helping themselves somehow. The crucial part is that admitting it carries no consequences — if people are afraid, you'll learn precisely nothing. In the companies where I've done this, it nearly always turned out two or three people were already using AI and nobody knew.
2. Draw the line around data, not tools. Banning "ChatGPT" makes no sense — stating what must never go into any external tool does. Typically: client personal data, contract contents, payroll and health records, credentials, non-public financials. That's a sentence an employee will remember. Nobody remembers a list of approved tools.
3. Give people an approved route. That could be a business tool with suitable contractual and security terms or an internal knowledge system over company documents. Check where each solution actually sends data; RAG alone does not guarantee that it stays in your infrastructure. In sensitive sectors, a model hosted on your own server may be worthwhile.
4. Keep a record of who uses AI and for what. This needn't mean monitoring — it's enough that every new tool is known about and signed off by someone. That record doubles as the documentation you'll need when somebody asks.
5. Train the staff, not just the managers. Not on "how AI works", but on what specifically must not leave the company, and why. Half an hour of plain explanation achieves more than a ten-page policy nobody finishes.
What not to do
Don't turn it into an investigation. The moment you start hunting for "who broke the rule", people close up and you lose visibility permanently. The goal isn't to find a culprit — the colleague in the opening had no offence in mind. The goal is to move that activity out of the grey zone and into the light.
Don't buy a "shadow AI detection" tool as step one. A market is already forming around this topic that will sell you software before you've even worked out what data you're protecting. A conversation with five people costs you an afternoon and tells you more.
Don't pretend it isn't happening. If anyone in your company works with text, spreadsheets or e-mail and you haven't given them a working alternative, the odds are high. The data is unambiguous: this isn't just a large-corporate phenomenon.
An honest conclusion
Shadow AI is an unusual kind of problem because it doesn't stem from negligence or bad intent. It stems from employees adopting new tools faster than companies can govern them. That isn't a failure of people — it's normal, and it was predictable.
The only unpleasant part is how expensive that gap can get. The good news is that it closes surprisingly cheaply: a conversation, one clear sentence about data, and one working alternative. None of those three requires a software budget.
If you take one idea from this article, let it be this: people will always find a route to faster work. The only thing you decide is whether it runs through your company or around it.
Frequently asked questions about shadow AI
As part of an AI audit I'll go through your processes and tell you where data moves, where the risk is and what to replace it with — concretely and without scare tactics.
Book a free 30-minute callI build AI solutions for businesses — from chatbots to knowledge systems. I write about what actually works, no buzzwords. More about me →