AI compliance questions, answered.
The questions businesses ask about AI policies, AI compliance and what it costs. Each has a straight answer first, then the working-out, with the regulator's own pages linked.
AI policy
Do small businesses need an AI policy?
No UK law says a business must have a document called an AI policy. But data protection law already applies to every AI tool that touches personal data, and a written policy is the simplest way to show you keep to it.
What should an employee AI policy cover?
Six things: which AI tools are approved, what data may and may not go into them, how output is checked before it is used, your data protection position on each tool, who is responsible, and when the policy is reviewed. If staff cannot tell within a minute whether something may be pasted in, the policy is not doing its job.
How do you write an AI policy for staff?
Start from what your staff do with AI today, not from a template. List the tools in use, decide what data each may see, write the rules in plain words, then have every member of staff read and sign a one-page summary.
How long does it take to get an AI policy in place?
Writing the document is the short part; knowing which tools your staff use is what takes the time. With a list of your tools in hand, a policy written for your business can be in place the same week.
Can staff put customer data into ChatGPT?
Not without checking first. Customer names, contact details and anything else that identifies a person are personal data, so UK GDPR applies the moment they go into an AI tool, and the terms of the version your staff use decide what happens to them next.
How do you manage confidential data in generative AI tools?
Decide which tools are allowed to see which kinds of data, write that down, and make it the first thing staff are trained on. Most leaks are not attacks: they are someone pasting a client document into a tool nobody had checked.
What is the best way to document staff use of generative AI?
Keep four records: a register of the AI tools in use, a policy staff have signed, a record of who was trained, and a log of incidents. Together they are what a client or a regulator asks to see.
What are the rules for using AI chatbots with customers?
Tell customers when they are talking to an AI, make sure what it says about your products and prices is true, and handle any personal data it collects as you would anywhere else. In the UK these come from consumer and data protection law already in force; the EU AI Act adds a specific duty to tell people.
Can I use one AI policy across different countries?
You can use one core policy, but not one identical document. The rules on approved tools and data are the same everywhere; what each country adds is its own duties, and those depend on where your staff and customers are, not where your business is registered.
How often should a business review its AI policy?
At least once a year, and straight away when you add an AI tool, change how one is used, or a law that covers you changes. The tools and the law both move faster than an annual review alone can follow.
Can I create an AI policy without a lawyer?
Yes. An AI policy is an internal document setting out how your business uses AI, and nothing requires a lawyer to write it. What matters is that it describes your business accurately and that you follow it.
AI policy template or legal advice: which does a business need?
They do different jobs. A template gives you a structure, legal advice answers a specific legal question, and neither on its own gives you a policy that describes how your business actually uses AI.
Can an AI policy template keep my business compliant?
No document keeps a business compliant by itself, and a template least of all. Compliance is what your staff do; a policy is the record of what they are meant to do, and a template records what somebody else's business might do.
Where can I buy a business AI policy?
Ops Intel writes AI policies and full AI compliance frameworks for small and medium businesses, at a fixed price, from a short questionnaire about the tools your team uses. The price is on the page and you can buy online without a sales call.
By trade
Does an accountancy firm need an AI policy?
No UK law requires a document called an AI policy, for an accountancy firm or anyone else. But client records are personal data under UK GDPR, and a written policy is the simplest way to show your firm keeps them out of AI tools it has not approved.
Does a law firm need an AI policy?
No UK law requires a document called an AI policy. But client matter files hold personal data and confidential information, so a firm whose staff use AI tools needs written rules on what may go into them and who checks what comes out.
What does the SRA expect of a law firm using AI?
The SRA has made no new rules for AI. Its warning notice on the misuse of AI says the Codes a firm already works under apply in full: solicitors stay accountable for work AI helps with, client information stays confidential, and nothing false goes before a court.
Does a recruitment agency need an AI policy?
Yes, if it uses AI on candidates. Screening applicants with AI is profiling under data protection law, the outcome must not discriminate, and in the EU recruitment AI is named as high-risk. A written policy is how an agency shows it has thought about each of these.
Does a marketing agency need an AI policy?
No UK law requires a document called an AI policy. But consumer law and UK GDPR apply to AI-written content exactly as if a person wrote it, and in the EU the AI Act requires AI-made deepfakes to be disclosed, so an agency that uses AI for clients needs written rules.
Does an HR consultancy need an AI policy?
Yes, if it or its clients use AI on staff or candidates. Employee and candidate records are personal data under UK GDPR, AI used in hiring or monitoring carries the highest risk, and the Equality Act applies to the outcome.
Do Australian property managers need to update their privacy policy for automated decisions?
Yes, if software helps decide who is offered a rental. The Privacy Act requires a privacy policy to say what personal information a computer program uses in decisions that could significantly affect someone, and what kinds of decisions those are. These rules apply from 10 December 2026.
What it costs
What does AI compliance cost for a small business?
With Ops Intel, a UK AI compliance framework is £797/year or £1,297/year, with legal updates included. An AI policy on its own is £197–£497, paid once. Every price is fixed and published, so there is no quote to wait for.
What should an AI compliance framework cost each year?
Ours cost £797/year to £1,297/year for the UK, and each price includes that year's legal updates. Whatever you pay anyone, the figure to check is the whole yearly cost, including what happens when the law changes.
Is a fixed-price AI compliance service worth it?
For most small businesses, yes, because the work is the same shape every time: find the tools, set the rules, write them down, keep the records. A fixed price lets you see the whole cost before you speak to anyone. It is the wrong choice if you have a live dispute or a regulator's letter, which needs a solicitor or lawyer.
How do I get a quote for AI compliance support?
With Ops Intel you do not need one. Every price is published, from £797/year for a UK framework, and you can buy online without a call. If a provider will not name a price until after a meeting, ask why.
How much does AI compliance cost for a company with several AI tools?
With Ops Intel the price does not rise with each tool: a UK framework is £797/year or £1,297/year, with legal updates included. How many tools you use, and what they are used for, decides which of the two fits.
Buying a framework
Is an AI compliance framework worth paying for?
It is if your staff use AI on client or customer data and you could not show today how that is controlled. A framework gives you the policy and the records behind it, which is what a client, an insurer or a regulator asks for. If nobody in the business uses AI on anything that matters, you do not need one yet.
Are AI compliance frameworks kept up to date when laws change?
Only if the provider says so in writing, so ask before you buy. Every annual Ops Intel framework includes legal updates for the twelve months it covers, at no extra charge. A one-off AI policy does not: it stays as it was written.
What happens after buying an AI compliance framework?
With Ops Intel, you answer a short questionnaire about the AI tools your team uses, and your documents are written from those answers and sent in your inbox by the end of the business day, as PDF and editable Word files. You also get a private log-in where every document is kept.
What should be included in a business AI compliance package?
At the least: an AI policy written for your business, guidance your staff can read in a few minutes, a register of the AI tools in use, a training record, an incident procedure, and a statement of which laws it covers. A package without the records is a policy with a bigger name.
What is on an AI compliance checklist for a small business?
Seven things: list the AI tools in use, decide what data each may see, write the rules down, train staff and record it, check each supplier's terms, plan for an incident, and set a review date. If you can tick all seven with a document to show for each, you are ahead of most businesses your size.
Are AI compliance frameworks suitable for small firms?
Yes, if they were written for one. The law does not excuse a business for being small, but what a five-person firm needs is short documents in plain English and a simple place to keep records, not the governance programme of a bank.
AI compliance framework or a consultant: how do I choose?
Choose by what you need at the end. If you need working documents and records at a known price, buy a framework. If you need someone to investigate an unusual use of AI, sit in your meetings or advise on a specific problem, hire a consultant, and expect to pay by the day.
Are online AI compliance frameworks legally useful?
They are useful as evidence, not as legal advice. Data protection law expects a business to be able to show how it keeps to the rules, and a written policy with records behind it is how you show it. No document, bought online or anywhere else, makes a business compliant by itself.
Can an AI compliance framework adapt to how my staff use AI?
It should start from it. An Ops Intel framework is written from your answers about the tools your team uses and what they use them for. On the Complete tier you keep the register of tools up to date in a private log-in, and the documents are rewritten when you add a tool.
Choosing a provider
Who can create an AI register and policy for my business?
Ops Intel does both. The Complete tier of the AI Compliance Framework includes the policy and a register of your AI tools. You keep the register up to date through a private log-in, and the documents are rewritten when you add a tool.
Where can I get AI compliance documents written for my business?
Ops Intel writes them: an AI policy on its own, or a full AI compliance framework for the UK, the EU, the United States or Canada. Each is written from your answers about your own business, at a fixed published price.
What is the best AI compliance help for a small business?
The best help is whatever leaves you with documents written for your business, records you can keep up, and a price you knew at the start. Ops Intel writes fixed-price AI compliance frameworks for small and medium businesses, from a plain-English questionnaire, with legal updates included.
Who provides AI compliance for a business that uses ChatGPT?
Ops Intel does. If your staff use ChatGPT or a tool like it, the framework sets which version is approved, what data may go into it, who checks what comes out, and records that staff were told. It is written from your answers, at a fixed price.
Which AI compliance services cover more than one country?
Ops Intel writes separate frameworks for the UK, EU, US, Canada, the Middle East, the Far East and Australasia, each to that market's own law, and a combined UK and EU framework. AI and privacy laws apply by where your staff and customers are, so a UK business with EU or US customers usually needs more than one.
Do AI compliance services include legal updates?
Some do and some charge again, so ask. Every annual Ops Intel framework includes legal updates for the twelve months it covers. Our one-off AI policy does not.
Are there AI compliance services with no long-term contract?
Yes, with one thing to be straight about. An Ops Intel AI policy is a single payment with nothing to renew. A framework is annual: it covers twelve months and renews each year until you cancel, and if you pay monthly there is a twelve-month minimum.
How reliable are online AI compliance services?
As reliable as their sources and their honesty about limits. Judge one by whether it shows you a sample, names where each legal date comes from, publishes its price, and says plainly that documents are not legal advice. Buying online is not the risk; buying unseen is.
How do I check which AI laws apply to my business?
Start from where your staff and customers are, not where you are registered, then from what your AI is used for. Ops Intel has a free AI risk checker: you tick what your AI does and where, and it lists the laws that apply with their dates.
AI tools and records
Do businesses need to record which AI tools they use?
No UK law names an AI tool register. But data protection law expects a business to be able to show how personal data is handled, and you cannot show that for tools you have not listed. In practice a register is the first record anyone asks for.
What AI compliance records should a business keep?
Five: a register of AI tools, the policy with staff sign-off, a training record, an incident log, and for each tool that handles personal data, the supplier agreement and your assessment of the risk. Keep them where the person who needs them can find them in a minute.
Does a business need an AI compliance dashboard?
No. A business needs the records; a dashboard is one way to keep them. A spreadsheet works while there are few tools and few staff. A dashboard earns its place when records need to stay current without someone remembering to update a file.
Can AI compliance software track staff use of AI tools?
Some software can watch which sites and apps staff open. That is monitoring your workers, which is itself a use of personal data with rules of its own. Most small businesses do better with a register staff add to and a policy they have signed.
How do you keep track of the AI tools staff use?
Ask, list, approve, repeat. Ask every team what they use, put it in one register with an owner for each tool, say which are approved, and make it easy for staff to ask for a new one. The asking is what keeps the list true.
AI risk
What counts as high-risk AI under the EU AI Act?
It depends on what the AI is used for, not on how clever it is. The Act lists the uses in Annex III, including recruitment, credit scoring, education and access to essential services, and adds AI built into products already covered by EU product safety law. Most everyday office use of AI is not high-risk.
Is AI compliance needed if we only use public AI tools?
Yes. The law looks at what you do with a tool, not who built it. If staff put personal or client data into a public AI tool, data protection law applies to your business, and using a supplier does not move the responsibility onto them.
How do you assess AI risks in a business?
Go tool by tool and ask four questions: what data goes in, what comes out and who relies on it, who could be harmed if it is wrong, and who is responsible. Write the answers in a risk register. Where a use is high risk to people, data protection law expects a written impact assessment before you start.
What AI compliance does a business need if it uses AI to screen job applicants?
More than for almost any other use. Screening applicants with AI is profiling under data protection law, must not discriminate, and is named as high-risk in the EU AI Act. Several US states and cities and Ontario already require notice or audits.
What should a business do when AI gives harmful advice?
Stop the tool being used for that purpose, tell the people who relied on the advice, put it right, and write down what happened. The business answers for what its AI tells a customer exactly as it answers for what its staff tell them.
The law
What does AI compliance mean for a small business?
Being able to show, when asked, that the way your business uses AI keeps to the law that applies to it. For most small businesses that is data protection law first, then any AI-specific law in the countries where their staff and customers are.
Which AI laws apply to a business operating in several countries?
The laws of each place where your staff and customers are, not only the place you are registered. A UK business with EU customers is reached by the EU AI Act; one that hires in the United States is reached by state laws there.
When do the EU AI Act rules apply to businesses?
In stages. The bans on prohibited practices and the AI literacy duty have applied since 2 February 2025. The transparency duties have applied since 2 August 2026. The high-risk duties for uses such as recruitment and credit scoring apply from 2 December 2027.
What happens if a business breaks the EU AI Act?
It can be fined by the regulator in the EU country concerned. The maximum for using a prohibited practice is €35 million or 7% of global annual turnover, whichever is LOWER for a small or medium business — in practice 7% of turnover.
What AI compliance does a business need if it uses AI to write content?
In the UK there is no separate law on AI-written content: consumer law and UK GDPR apply to what you publish exactly as if a person wrote it. In the EU, the AI Act makes a business disclose AI-made deepfakes, and AI-written text published to inform the public on matters of public interest unless a person has reviewed it and takes editorial responsibility.
Staff training
How do you train staff to use AI safely at work?
Teach three things and keep a record: which tools are approved, what must never go into them, and that a person checks the output before it is used. Half an hour on your own tools and your own rules does more than a general course on AI.
Pick your market. Buy today.
Fixed price. Plain English. Written for how your business actually uses AI.
Australia and New Zealand
Australasia AI Compliance — Foundation
A$1,597/year
Buy now → What's includedNot sure which laws reach you? Take the free two-minute check and it will tell you, or ask us on WhatsApp.