The chat on our website is not AI#
The plan builder and the question panel on our website are rule-based: they follow a fixed script and a fixed calculation, and they run in your browser. No AI model reads or answers what you type there, and your answers are not sent to us unless you continue to registration (see the Privacy Policy). If we ever connect an AI model to the website chat, we will label it clearly as an AI assistant before you use it.
How recommendations are calculated#
When the Service recommends an advertising platform, a daily budget and an estimated range of weekly leads, it uses fixed rules, not an AI model:
- the platform is chosen from those suited to your goal and connected accounts;
- the daily budget comes from the budget in your brief, from your target cost per lead, or from a standard starting budget for your goal, and never below the platform's minimum;
- the estimated leads are the budget over a week divided by an expected cost per lead, shown as a range of plus or minus 30 %.
Every figure is an estimate, not a promise. The reasons are shown next to each recommendation, and nothing is launched until you approve it.
Where the dashboard uses AI models#
Features marked as AI in the dashboard use large language models and image models to propose content: site pages and texts, ad texts, social posts and images. They use AI tokens from your plan.
- Provider: [to be completed: AI model provider].
- What is sent: the brief, product information and instructions you use for a generation. Lead data is not sent to AI models.
- Training: [to be completed: whether the AI provider may train on your inputs].
- Your control: AI output is a proposal. You can edit it, regenerate it or discard it, and nothing is published until you approve it.
Limits of AI output#
AI output can be wrong, out of date, biased or similar to content made for others. Check facts, prices, legal and health claims before you publish, and make sure the result follows the advertising rules that apply to you. Do not enter passwords, payment details or sensitive personal data into prompts.
Labelling#
Where the law requires it, for example the EU AI Act, we tell you when you are interacting with an AI system, and content generated or edited by AI can be marked as such.
Contact#
To talk to a person about any of this, write to [to be completed: contact email].