What is a private AI assistant?

An assistant that answers questions from your own documents, such as policies, procedures and handbooks, and shows where each answer came from. It only shows people what they already have permission to see. The technique is often called retrieval-augmented generation, or RAG, and our article on RAG in plain English explains it.

Is it safe to put our business information into AI tools?

It depends on the tool and how it's configured. Business editions of the major AI assistants offer much stronger data protections than free consumer versions. We help you choose appropriate tools, configure them properly, and set clear guidance for your team, with UK GDPR in mind. Your data is never used to train models, and you can exclude any system or data from AI handling.

Which AI tool should we use?

It depends on what you already use, what you want to achieve, and your data protection needs. If you're on Microsoft 365, Copilot may be a natural fit, but it isn't always the right answer. We know Copilot well and can supply Microsoft licences, but we'll recommend whatever suits you best.

Will AI replace our staff?

That's not how we approach it. We focus on taking repetitive, low-value work off people's plates so they can spend more time on the work that needs their judgement and relationships.

AI and automation for business operations

Is this for you?

You'd like staff to get reliable answers from your policies, procedures or contracts without searching through folders
Some staff are already using AI tools informally, and you'd like some consistency and guidance
People re-type or copy information between systems, or turn emails and forms into records by hand
You'd like to use AI on your own information without it being used to train anyone's models
You want to move at a sensible pace, with a clear view of the value

What we do

Private AI assistants. Assistants that answer questions from your own documents and point to the source, built so that people only see what they are already allowed to see. Our article on RAG in plain English explains the technique.
Automation. Small, well-defined automations, for example turning emails or forms into structured records or drafting routine documents, with a person approving anything consequential.
AI features in your software. Adding assistants and automation to applications we build or look after, tested like any other part of the system.
Choosing and setting up tools. Straight advice on which AI assistant fits your organisation, then careful set-up, including permissions, data protection settings and licensing.
AI use policy. A short, practical policy your team will actually read: which tools to use, what information can go in, and how to check outputs.
Getting your information ready. Tidying the files, folders and permissions that AI tools rely on, so they give better answers.

The same controls as software care

We use AI in our own engineering, and we apply the same method to the AI and automation we build for you.

Specification first. We agree what the assistant or automation should and should not do before we build it.
Tested. Automated tests describe the expected behaviour, so changes can be checked rather than trusted.
Independent review. A separate reviewer checks every change.
Human sign-off. A named engineer approves anything that reaches a live system.
Your data stays yours. Customer data is never used to train models, and you can exclude any system or data from AI handling.
Your repository. Code and documentation stay in your own repository, so you are never locked in.

Once it is live, software care can look after it with a named engineer accountable for it. The method is described in more detail in how we work.

How it works

Conversation. We understand your organisation, your current tools and what you would like AI or automation to help with.
Options. We recommend where it would help most, and say plainly where a simpler approach would do better.
Pilot. We build a small version with a few people and a few real tasks, with the controls above in place.
Roll out and support. We extend what works, document it, and keep it on track as the tools change.

Practical AI and automation that earns its place.

Is this for you?

What we do

The same controls as software care

How it works

Frequently asked

Further reading

AI agents explained: what they are and where to start

Connecting the apps you already use: small automations that add up

RAG in plain English: how AI can answer questions from your own documents

Let's talk about what's next for your business.