AI terms and conditions
Our service is delivered using AI and automation, with human oversight. These terms explain how we use AI, what you can expect from AI-made work, and your responsibilities when you use it. They form part of our terms and conditions.
Last updated: 24 October 2026
1. Words we use
- AI means artificial intelligence tools, including large language models, that create text, images, audio or replies.
- Input means anything you give us or put into our AI tools, such as business information, product details, photos, documents and website content.
- Output means anything our AI tools create for you, such as articles, posts, ads, review replies, newsletters, email automations, magazine pages, listings, website pages, chatbot replies, press releases, animated images, podcast episodes, slides and competitor reports.
2. How we use AI
Our bots do most of the work: they write, check and suggest. People oversee the work. Where a plan includes approvals, you can check work before it goes out:
- Changes to your ads are applied with your approval, or we just get on with it if you prefer: your choice.
- Review replies can be set to “approve first”.
- You can approve each email first, or let it send automatically.
- Press releases go out only after you approve them.
3. Our AI providers
We use AI models from third-party AI providers, such as OpenAI, Anthropic, Google, Microsoft and xAI. They process Input to create Output on our behalf, under their business terms. AI processing may happen in the USA. Where it does, it is protected by the UK–US data bridge (for providers certified under it) or by the UK International Data Transfer Agreement (IDTA) or the UK Addendum to the EU Standard Contractual Clauses.
4. Your data and AI training
We do not use your data to train our own models. Our AI providers handle Input under their own terms and policies, which govern how they use and keep it. We try to limit the personal and confidential information we put into AI tools to what is needed for the work. If any of your material must not be processed by AI tools, please tell us.
Our privacy policy and data processing agreement explain how we handle personal data.
5. AI is not perfect
AI can get things wrong. Output may sometimes be inaccurate, out of date, incomplete or similar to content made for others. We check work, but:
- Output is not professional, legal, medical or financial advice;
- you should check facts that matter, such as prices, opening hours and claims about your products, before you approve or publish;
- we only promise what we make ourselves (articles, posts, listings, pages, checks and reports), never rankings, links or traffic.
If Output we deliver is wrong, tell us and we will correct it.
6. Powerful Bots: your chatbot
If you have a Powerful Bots plan, your chatbot is installed on your own website with an embed code. It answers your visitors using the business information and content you give us.
You agree to:
- make it clear to your visitors that they are talking to an AI chatbot, not a person;
- mention the chatbot in your own website privacy policy;
- keep the information your bot uses accurate and up to date, and tell us when it changes;
- not use the bot to collect special category data (such as health information) or card details.
Your chatbot runs on Kuga, our chatbot platform provider, which processes data in the UK, EU or USA under appropriate safeguards. Chatbot data is deleted within 14 days after your plan ends.
7. Podcasts and Images
- Powerful Podcasts use AI voices, and the show notes say so. We do not clone anyone’s voice or edit your own recordings.
- Powerful Images are made only from photos you own or are licensed to change. We do not animate photos of people without their consent, and we label AI-made images where a platform asks for it.
8. Chatting with a bot on our website
If we offer a chatbot on our website and you use it, you are talking to AI, not a person. Please do not share sensitive personal information in the chat. We use chat messages to answer you, as described in our privacy policy.
9. What you must not do
You must not use our AI services, or ask us, to:
- create anything unlawful, harmful, defamatory, discriminatory or misleading, or that breaks someone else’s rights;
- pretend to be another person or business, or copy someone’s voice or likeness without permission;
- create fake reviews, or reviews posing as customers;
- break the rules of Google, Meta, eBay, Amazon, Etsy or any other platform;
- try to get round our tools’ safety controls, or copy or reverse-engineer our bots and prompts.
We may refuse or remove work that breaks these rules.
10. Ownership
- Input: you keep ownership of your Input. You give us permission to use it only to deliver your plan.
- Output: once you have paid for it, you can use the final Output we deliver for your business. You may not resell Output on its own (for example, selling our articles to others) without our written permission.
- Our tools: our bots, prompts, templates and processes remain ours.
11. Changes
AI tools change quickly. We may change the AI providers or models we use to keep the service working well. We will update this page when we change providers, and tell clients by email where our data processing agreement requires it.
