The Hidden Risks of Using Public AI Tools at Work

Public AI can save you hours, however are you sharing more than you realise?

Artificial Intelligence has quickly become part of everyday business. Employees are using tools like ChatGPT, Gemini and Claude to draft emails, summarise documents, write reports, brainstorm ideas and analyse information. Used correctly, these tools can save hours of repetitive work and significantly improve productivity. There’s one important question every business should ask before using public AI tools:

Would you be comfortable sending this information to a complete stranger?

If the answer is no, it probably shouldn’t be entered into a public AI platform. The reality is that many SMEs are unknowingly exposing confidential business information simply because they don’t understand the difference between public AI and secure business AI solutions.

Future-proof your business with secure, private AI. E-DATA helps SMEs design, build, and deploy their own private AI systems, ensuring you get the benefits of advanced technology without compromising your data security. Ready to see what AI can do for you? Visit our contact page to connect with an advisor and explore our funded support.

Want to learn more about AI? Read on.

What Are Public AI Tools?

Public AI tools are the freely available services anyone can access through a web browser or mobile app. Examples include:

ChatGPT
Google Gemini
Claude
Perplexity

These tools are incredibly powerful and have transformed how many people work. However, they operate outside your organisation’s own systems and policies. While many providers now offer privacy controls and enterprise subscriptions, businesses should always understand where their information is going before uploading sensitive data.

What Should Never Be Uploaded?

One of the simplest rules is this:

If you wouldn’t post it publicly online, don’t assume it’s suitable for a public AI tool.

Examples include:

Customer information
Financial reports
Contracts and legal documents
Employee records
Product designs
Source code
Pricing information
Commercial agreements
Business strategies
Confidential meeting notes

Many employees upload this type of information simply because they’re trying to save time not because they’re trying to create a security risk.

The Risks Businesses Often Overlook

Confidential Business Information
Your business information is valuable. Uploading internal reports, proposals or pricing models into public AI tools may mean you are sharing commercially sensitive information outside your organisations control. Even if no breach ever occurs, many organisations have contractual obligations to keep this information confidential.
Customer Data
Customer trust is one of your greatest assets. Uploading names, addresses, contact details or customer records into public AI tools could expose personal information in ways your customers never expected. Before using AI with customer information, businesses should always consider whether that data has been appropriately protected.

Intellectual Property
Many SMEs use AI to help write technical documentation, improve products or brainstorm new ideas. What if those ideas form part of your competitive advantage? Product designs, engineering drawings, research findings and innovative concepts are all valuable intellectual property. Protecting that information should always come before convenience.

GDPR and Compliance
Many organisations are subject to legal and regulatory requirements around how information is stored and processed. If personal data is uploaded into an AI platform without appropriate safeguards, businesses may struggle to demonstrate compliance with UK GDPR and other contractual obligations. Understanding where your data is processed and who has access to it is becoming an increasingly important part of responsible AI adoption.

AI Isn’t the Problem, How We Use It Is

Public AI tools are incredibly useful for tasks such as:

✔ Drafting emails
✔ Brainstorming ideas
✔ Creating meeting agendas
✔ Improving grammar
✔ Generating marketing content

These are generally low-risk uses because they don’t require confidential business information. The risk arises when organisations begin uploading sensitive documents without considering the potential consequences.

What Are the Alternatives?
Fortunately, businesses don’t have to choose between productivity and security. Many organisations are now exploring private AI environments that allow them to benefit from generative AI while keeping their information securely within their own systems.

Instead of sending information to a public AI platform, a local Large Language Model runs entirely within your own secure environment. Your information stays under your control, making it a suitable option for organisations working with confidential or commercially sensitive data. Rather than asking a public AI model to search the internet, E-DATA can support you with AI systems to search your own documents securely. Imagine being able to ask questions like:

“What does our employee handbook say about annual leave?”
“Find our latest health and safety policy.”
“Summarise this project documentation.”

AI provides answers using your own trusted documents, reducing the risk of inaccurate information while keeping your data private.

How Can SMEs Use AI Safely?

You don’t need to ban AI from your organisation. Instead:

✔ Create clear guidance for staff.
✔ Decide what information can and can’t be entered into public AI tools.
✔ Consider secure AI solutions for confidential business information.
✔ Train employees to use AI responsibly.
✔ Review your existing data protection and information security policies to include AI.

AI should become part of your business toolkit, not a hidden source of risk.

How E-DATA Can Help

Artificial Intelligence is developing rapidly, and every business is at a different stage of its journey. Whether you’re exploring AI for the first time or considering more advanced solutions such as local Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), workflow automation or secure AI deployments, E-DATA can help you identify the right approach for your business.

Through our funded workshops, mentoring and implementation projects, we help SMEs understand where AI can deliver real value while ensuring solutions remain practical, secure and aligned with your business needs.

If you’re interested in exploring how AI could support your organisation safely and effectively, visit our Contact Us page and one of our advisors will be happy to discuss the support available through E-DATA.