An AI employee is a role-specific AI system that does a defined job inside your business. It reads what comes in, decides what to do, acts in the tools you already use, and reports back, with a person approving anything that matters. Assist IQ, a Scottish AI automation agency, builds and manages AI employees and agents for small businesses across Scotland and the UK.
Most businesses have used AI in one of two ways. They have asked ChatGPT to write something, or they have set up an automation that moves data from one place to another. Both are useful. Neither is an AI employee.
Most of it is a chatbot in a suit, or an automation with a language model bolted onto one step. The industry calls it agent washing. The test for a real AI employee is simple, and it has three parts.
The difference is not the AI model underneath. It is whether the system has been connected to the business properly: real context, real tools, a real scope, and a human where judgment is required.
| Chatbot | Automation | AI employee | |
|---|---|---|---|
| What it does | Answers questions when asked | Runs a fixed sequence when a trigger fires | Holds a role: monitors, decides, acts and reports within limits you set |
| How it decides | Responds to what you type | Same path every time, no judgment | Reads the context and chooses the right action for this situation |
| What it knows | Public training data | Whatever fields are mapped | Your documents, your history, your rules, your live systems |
| When it acts | When you ask | When the trigger fires | Continuously, without being prompted |
| Who is accountable | Nobody. It answered. | Whoever built the workflow | A named role with a scope and an escalation path |
These are not wireframes or capability demos. They are the businesses we run and the systems we built for them. The full stories are on our work page.
ScottishWill
The operator that runs a will-writing business
A guided journey gathers what is needed, automations carry the admin, and no AI ever touches the legal document itself. One person runs what would otherwise take four. The business would not be viable without it.
HALO Intervention Services
Paper-based operations made compliance-grade
An oil and gas well intervention firm moved off paper onto processes designed around AI automation and augmentation. The entire ISO 9001 management system was generated on the same principles.
PCMG Property
A property business that does its own paperwork
Fully automated expense management and zero-touch compliance: insurance chased, gas certificates chased, nothing waiting on someone to remember. Passive income, once it is actually systemised.
Honest answer: it depends on the role, so we price after we understand your business, not before. The ladder below is how every engagement starts. The first rung is the only thing on this page with a price.
Straight answers. More on the FAQ page, and the review itself is where we answer the questions specific to your business.
An AI agent is the technology: a system that can read inputs, use tools and take actions. An AI employee is that agent given a proper job. A defined role, a scope it stays inside, checks on its output and a person accountable for it. Every AI employee we build runs on agents underneath. The employee framing is what makes it safe and measurable in a real business.
Yes, if the role is well defined. The businesses that struggle with AI started too broad. The ones that succeed start specific: every missed call handled, every quote chased until it is a yes or a no. A single well-scoped role costs less than the admin time it replaces, and it works around the clock. The AI Ops Review exists to find the roles that pay for themselves first.
A single well-defined role typically goes live within days of agreement, not months. Bigger builds, such as a BusinessOS system built around your own documents and knowledge, are scoped in weeks. There is a limit to how fast this should go: the system has to learn how your business works and pass its checks before it touches a real customer.
Not in the way the question usually means. The businesses running AI employees well are not cutting heads. They are expanding capacity: the same team handles more enquiries, more clients and more paperwork because the system carries the repeatable layer. Your people keep their jobs and lose the part of the job they hate. Judgment stays with a person.
The same thing that happens when any employee makes a mistake: you find out, you understand why, and the system gets fixed. The difference is that an AI employee makes mistakes in patterns, so you fix the pattern once. Every system we build has checks on its output, human approval for anything consequential, and an escalation path for when it is not sure.
Not a discovery call. Not a pitch with a calendar link. The AI Ops Review is 45 minutes of questions about how your business actually runs. Then you get a written plan of what’s worth doing, what it costs, and what isn’t worth doing at all.
It’s £750. If we can’t find you at least 5 hours a week, it’s free. If you build with us inside 60 days, the £750 comes off the bill. So the only way it ever costs you money is if you take the plan and run it yourself. Fair enough.
Worst case, you lose 45 minutes and leave with straight answers.
No prep. No data handover. No long contract. You talk, we listen.