Shadow AI is when staff use unapproved AI tools at work, often without realising the risk. It can expose client data, breach GDPR, and widen your attack surface. To prevent the risks of shadow AI, create a policy around AI tool usage, train your staff on approved AI tools, and audit regularly.
What Is Shadow AI?
You have probably heard of shadow IT. Someone installs unapproved software or signs up to a cloud tool without telling anyone. Shadow AI is the same idea, but the risks go further.
Shadow AI refers to the unsanctioned use of artificial intelligence tools by employees, without IT knowledge or sign-off. Gen AI tools are free, fast, and genuinely useful, which is exactly why they spread.
According to a Microsoft survey of UK workers, 71% have used consumer AI tools at work without IT approval, yet most firms have no AI governance in place at all.
Unlike shadow IT, the risk is not just about where your data sits. Generative AI models are cloud-based, meaning anything your staff put in gets processed on external servers, outside your control and can even be retained long after the conversation ends.
What Does Shadow AI Actually Look Like in a Law Firm?

Rushed document drafts
A fee earner copies an internal strategy deck into a free AI tool before a client call. That company data is now sitting on an external server, outside your control.
Browser extensions and AI plug-ins
Someone installs an AI writing assistant as a plug-in. It reads everything they type, including client emails and case notes, often well beyond what the user expected.
Personal account workarounds
According to Menlo Security, 68% of employees used personal accounts to access free AI tools in 2025, with 57% entering sensitive data in the process. Restrict access at work and people find another way, with none of your firm’s security controls in place.
AI features embedded in existing tools
AI capabilities are being built into tools your staff already use, quietly activated through updates. Features embedded in a personal Microsoft account sit entirely outside your security oversight, even if the tool looks familiar.
What Are the Significant Risks for Law Firms?
Data leaks and client confidentiality
Unauthorised AI use means customer data can enter systems you do not control. According to a RiverSafe survey of UK CISOs, 1 in 5 UK companies has already experienced data leakage through employee use of generative AI.
GDPR and SRA compliance violations
Adopting AI tools without proper vetting can put your firm in breach of GDPR. The SRA takes data security seriously, and the EU AI Act is adding further obligations on top.
Inaccurate AI generated outputs
Unvetted AI apps can produce AI generated content that looks authoritative but is simply wrong. In a legal context, an AI-generated draft that contains errors could influence advice or reach a client completely unchecked.
A weaker security posture
Personal accounts have no enterprise security controls, no MFA enforcement, and no visibility from your IT team. By the time most firms know which AI tools are actually in use, a security incident has already forced the question. You can read more about how Labyrinth Technology approaches network security here.
How Do You Mitigate Shadow AI?

Write a policy and make it usable
Define which AI tools are approved, what data can go into them, and how AI generated outputs must be reviewed. Ensure it is brief and simple to understand for everyone in the organisation.
Give people sanctioned AI tools that work
People use unauthorised AI tools because they enhance productivity and there is nothing better on offer. Providing enterprise-grade alternatives removes the incentive and gives security leaders full visibility over AI use. If you are not sure where to start, our digital transformation strategies service helps firms adopt AI properly, with the right governance from the beginning.
Train your staff
Security awareness training is not just for phishing. Your people need to understand what shadow AI is, the AI risk it creates, and how to request approval for tools they want to use.
Audit regularly
New AI tools appear constantly, and the ones your staff are using today may not be the same ones they are using in six months. Regular audits, led by your IT team or provider, keep your approved list current and your exposure visible.
How Labyrinth Technology Can Help
Labyrinth Technology has been supporting law firms in London for over 20 years. We help firms get visibility over unauthorised AI use, build realistic governance frameworks, and put the right cyber security controls in place before shadow AI becomes a serious problem.
Shadow AI is not a future concern. It is happening now. Get in touch for a free consultation and let us help you get on top of it.









