Skuto AI Policy Builder
Draft a plain AI-use policy for your team in a few minutes. It also makes an AI-literacy training record and an "I used AI" disclosure. Runs in your browser.
Answer a few questions
Free · Nothing is uploaded · The draft updates as you type
Leave it empty to get a [blank] you fill in later.
Separate them with commas, for example DeepL, Grammarly.
A name or a role, for example "Anna" or "the office manager".
The European Commission says the AI Act also reaches organisations outside the EU when their AI is used in the EU or affects people there. "Yes" adds a section on the AI literacy duty.
Your draft
Gaps to fill. Fill in: who approves new tools. Parts in [brackets] are blanks for you to fill in.
# AI-use policy: [Team or company name] _Draft of [date]. A starting point to agree with your team, not legal advice._ ## 1. Why we have this AI tools help us work faster. This page makes sure we all use them the same safe way: which tools, what never goes in, who checks the result, and who to ask. ## 2. Who this covers This covers everyone who works for [Team or company name], including freelancers and contractors who handle our files. ## 3. Which AI tools we use Approved AI tools: ChatGPT. Use them only with a work account that [Team or company name] set up, never with a personal account. Any other AI tool needs approval first (see section 7). That includes browser extensions and new AI features inside apps we already use. ## 4. What never goes into an AI tool Never paste, upload or dictate these into any AI tool, approved or not: - Personal data (yours or someone else's) - Client or customer data - Health information - Financial details - Passwords and keys - Company internal information Not sure? Check the text first with the free Paste Checker (https://skuto.app/tools/paste-checker/). When in doubt, leave it out or replace names and numbers with placeholders. ## 5. Check before you use it The person who uses AI owns the result. Before it leaves your hands, read all of it, check facts, numbers, names and links against a real source, and make sure it says what you mean. AI can sound sure and still be wrong. Anything that goes to a customer, gets published or affects a decision about a person is also read by a second person first. ## 6. Saying when AI helped Inside the team, no label is needed. Work that goes to customers or the public carries a short note when AI helped make it, such as "Drafted with AI, checked by [name]". Never present an AI-made image, voice or video of a real person as real. ## 7. New tools and questions Want to try a new AI tool? Ask [name or role] first. They check what happens to the data you put in and, if it is fine, add the tool to section 3. Questions about this policy go to [name or role] too. ## 8. Review We review this policy every 12 months, and sooner when we add a new AI tool. Next review: ________. Agreed by: ________ Date: ________
This is a starting draft, not legal advice. Check your school's or employer's own rules before you use it.
Built in your browser from templates. No AI writes it, and nothing you type is sent anywhere.
Not sure if a text is safe to paste? Check it with the Paste Checker.
Picking a team plan for these tools? Compare plans in the Plan Picker.
Last verified:
How it works
- Pick what you need: rules for your team, a training record, or a note saying you used AI.
- Answer a few short questions. An example draft is there from the start and changes as you answer.
- Copy it, download it or print it as a PDF, then edit it with the people it is for.
What goes in a small-team AI policy
A good AI policy for a small team fits on one or two pages. It answers the questions people already ask each other, so everyone can say yes or no without waiting for a meeting. The draft covers six things:
- Which AI tools are approved, and whether work or personal accounts are fine.
- What never goes in: personal data, client data, health and money details, passwords, internal files.
- Who checks the result before it reaches a customer or the public.
- When to say AI helped, with a short label people can copy.
- Who approves a new tool, so nobody signs up to something on their own.
- When you look at it again, because the tools change every few months.
The "never goes in" list uses the same groups as our Paste Checker, so your team can test a text against the policy before they paste it. Choosing a team plan for the tools you approve? The Plan Picker compares them.
The EU AI-literacy duty, in plain words
If your organisation uses AI systems in the EU, Article 4 of the EU AI Act asks you to help your staff, and anyone using AI on your behalf, understand the AI they work with. It has applied since 2 February 2025. The Commission says the rules on supervision and enforcement apply from 2 August 2026.
The article was rewritten by Regulation (EU) 2026/1744 (Digital Omnibus on AI), published in the EU Official Journal on 24 July 2026. Here is the current text, word for word:
Article 4: AI literacy
1. Providers and deployers of AI systems shall take measures to support the development of AI literacy of their staff and other persons dealing with the operation and use of AI systems on their behalf, taking into account their technical knowledge, experience, education and training and the context the AI systems are to be used in, and considering the persons or groups of persons on whom the AI systems are to be used. This obligation does not require providers or deployers to guarantee any specific level of AI literacy of any individual.
2. The Commission and the Member States shall support and facilitate the efforts of providers and deployers of AI systems, in particular SMEs, in fulfilling their obligation under paragraph 1 of this Article. For that purpose, the Commission shall publish practical examples of how to comply with that obligation on the single information platform referred to in Article 62(3), point (b).
3. The Board shall adopt recommendations, taking into account European competence frameworks, to support the Commission and Member States in the promotion of AI literacy required under paragraph 1, including by setting out common objectives.
The European Commission answers the common questions in its AI literacy Q&A (updated 20 October 2026). The points that matter for a small team:
- "There is no need for a certificate. Organisations can keep an internal record of trainings and/or other guiding initiatives."
- Since the Digital Omnibus on AI, AI literacy is still a duty for providers and deployers, but no specific or "sufficient" level is required.
- A company whose staff use ChatGPT for tasks like writing ad copy or translating is covered too: the Commission says they should be told about the specific risks, for example hallucination.
- As a minimum, the Commission suggests: a general understanding of AI (what it is, how it works, what your organisation uses, its opportunities and dangers), your role (do you build AI or use someone else's), the risks of the AI you use, and training fitted to what each person already knows and how they use AI.
- There is no set format: training, guidance and other initiatives can all count. The Commission adds that, in many cases, just asking staff to read a tool's instructions for use may not work.
Source: AI Literacy: Questions & Answers (European Commission) (checked 30 October 2026)
For a small team, that can be a short session fitted to the tools you actually use, plus a simple note of who learned what and when. The training record in the tool gives you that note, with the law quoted at the top.
Based outside the EU? The Commission says the AI Act still reaches you when your AI system is used in the EU or affects people there. If neither is true, the same record is still a simple way to show a client or a school that your team knows how to use AI safely.
How to say you used AI, honestly
Saying you used AI is the opposite of cheating: it tells your teacher, client or readers exactly where the help came from. A good note has three parts:
- The tool you used, by name.
- What it did: ideas, editing, translation, first drafts or code.
- What you did yourself: checked the facts, rewrote it in your own words, tested the code.
The rules are not the same everywhere. Schools differ, and so do courses within one school: some want a note at the exact spot, like a citation, others want a list or a signed declaration at the end. Clients and publishers have their own rules too. The tool gives you three versions, so you can match the one that is asked for. If a task says no AI at all, a note does not change that.
Nothing you type leaves this page
The drafts are put together in your browser from fixed templates. No AI writes them, and nothing you type (names, your team, titles) is sent to Skuto or to anyone else, or kept after you close the tab. Like the rest of Skuto, this page runs site analytics if you accept them: they learn which of the three drafts you made and whether you copied, downloaded or printed it, never the words. Session recordings blank the tool area.
What this tool cannot do
- It does not give legal advice or tell you whether your organisation is compliant. It gives you a draft to check and adopt.
- It does not know your school, employer or client rules. Those win over any draft.
- It does not judge whether a use of AI is allowed. It only helps you describe it truthfully.
- It does not cover high-risk AI systems, which carry extra duties under the AI Act.
Frequent questions
What should a small business AI policy include? +
Six things: which AI tools are approved and with which accounts, what must never be pasted into them, who checks AI work before it goes out, when to say AI helped, who approves a new tool, and when the policy is reviewed. One or two pages is enough for most small teams. A policy nobody reads protects nobody.
What does Article 4 of the EU AI Act require? +
Providers and deployers of AI systems must take measures to support the AI literacy of their staff and of other people who use AI systems on their behalf. It has applied since 2 February 2025. Since Regulation (EU) 2026/1744 (Digital Omnibus on AI), the text also says this does not require any specific level of AI literacy. The Commission says the rules on supervision and enforcement apply from 2 August 2026.
Do I need a certificate for AI literacy training? +
No. The European Commission's AI literacy Q&A says: "There is no need for a certificate. Organisations can keep an internal record of trainings and/or other guiding initiatives." The training record in this tool is a simple version of that internal record.
Does the AI literacy duty apply if my staff only use ChatGPT? +
A company whose staff use ChatGPT for tasks like writing ad copy or translating is covered too: the Commission says they should be told about the specific risks, for example hallucination.
How do I say I used AI in an assignment? +
Name the tool, say what it did (ideas, editing, translation, drafting or code) and say what you checked or changed yourself. Then follow your school's format: some want a note right where you used AI, others want a list or a signed declaration at the end. If a task does not allow AI, a statement does not make it allowed.
Is what I type sent to an AI or saved? +
No. The drafts are built from fixed templates in your browser. No AI writes them, nothing you type is sent to Skuto or anyone else, and nothing is kept when you close the page.
Is this legal advice? +
No. It gives you a clear starting draft. Your employer, school, client or a lawyer may need something different, so check their rules before you adopt or hand it in.
Tools you might need next
- Free Skuto Plan Picker Find out in one minute whether a paid AI plan is worth it for you: real prices, each with its source.
- Free AI for Beginners Brand new to AI and not sure where to start? Answer a few simple questions and we'll point you to the one assistant that fits what you want to do: honest, with sources, no hype. Start here.
- Free Skuto Paste Checker Check before you paste: pick what you're about to share and which AI you use, and see in seconds if it's safe, with the vendor's actual terms and a safer alternative.