Authority explainer · last reviewed 18 November 2026
What is shadow AI?
This page is maintained by the instructor of the Udemy course linked below. Figures below cite the report they are drawn from, checked 18 November 2026.
Short answer
Shadow AI is staff using AI tools their employer has not approved, provided, or even noticed, often a personal chatbot account standing in for one the company never set up. It shows up because people want a fast answer, not because they are trying to break a rule. A 2024 Microsoft and LinkedIn study found most workers who use AI at work bring their own tools, and a 2025 IBM study found organisations with a high level of shadow AI face noticeably higher breach costs than those with low or none.
What counts as shadow AI
Shadow AI covers any AI tool an employee uses for work that nobody in the organisation signed off on or knows about: a free chatbot account opened with a personal email, a browser extension installed without asking, or a team subscription nobody logged. It is not usually an attempt to get around a rule. Most people reach for it the same way they would reach for a personal phone to look something up: it is already open, it is faster than asking, and it works well enough.
How common is shadow AI
Three in four knowledge workers (75%) already use AI at work, and 78% of those users bring their own AI tools to work, per Microsoft and LinkedIn's 2024 Work Trend Index. That gap, between how many people are already using AI and how much of it an employer actually provided, is the space shadow AI fills.
Microsoft, 2024 Work Trend Index on the state of AI at work
What shadow AI costs
IBM's 2025 Cost of a Data Breach Report defines shadow AI as workers downloading or using unapproved internet-based AI tools. Organisations with a high level of shadow AI saw USD 670,000 added to their average breach cost compared with those with low or no shadow AI. The same report found 63% of the organisations studied by the independent Ponemon Institute had no AI governance policy to manage or restrict shadow AI use.
IBM, 2025 Cost of a Data Breach Report: navigating the AI rush without sidelining security
What reduces shadow AI at work
Naming which AI tools staff can use for which tasks, teaching people what each tool does with the text and files pasted into it, and offering an approved option good enough that nobody needs to look elsewhere all shrink how much work ends up running through accounts nobody chose. Training works best when it comes before the first personal account gets opened, not after.
Frequently asked questions
What is shadow AI?
Shadow AI is staff using AI tools at work that their employer has not approved, provided, or even noticed, usually because an approved option is missing, slow, or does not exist for the task at hand.
Is shadow AI the same as BYOAI?
They overlap. BYOAI (bring your own AI) describes an employee supplying their own AI tool for work; shadow AI is the wider pattern of any AI use outside an employer's visibility, whether that is a personal chatbot account, a browser extension, or a tool a team adopts without telling anyone.
Why does shadow AI happen even where a workplace policy says not to?
People reach for whatever finishes the task fastest. If the approved tool is missing, hard to reach, or plainly worse than a free consumer option, staff route around it rather than wait, especially under deadline pressure.
What reduces shadow AI at work?
Naming which AI tools staff can use for which tasks, teaching people what each tool does with the text and files pasted into it, and making the approved option good enough that nobody needs to look elsewhere.
Sources
- Microsoft, 2024 Work Trend Index on the state of AI at work, 8 May 2024
- IBM, 2025 Cost of a Data Breach Report: navigating the AI rush without sidelining security
General information, not legal advice. Last reviewed 18 November 2026.
Give staff a sanctioned place to point their AI use.
A 45-minute workplace baseline on Udemy covers how AI tools handle data and when to ask before pasting something in.