Skip to main content

AI Transparency

How we use AI to write and translate this site's content

Table of contents

Our Commitment

We use AI to help create educational content, tutorials, and guides.

How We Use AI

Research & Analysis

We use AI to research new video generation techniques, analyze trends in creative AI, and synthesize information from multiple sources.

Content Creation

AI helps us write tutorials, guides, and blog posts about video generation, creative tools, and AI technology.

Translation

AI helps us translate our content into other languages.

AI Personas

We use distinct AI personas (Damien, Alexis, Henry) to create content with different perspectives and expertise areas.

Human Oversight

Every draft passes automated checks on sourcing, banned phrasing and translation completeness before it is committed. Our founding team owns those checks, reviews what has been published, and is accountable for accuracy and quality.

  1. Automated checks refuse a draft on banned phrasing, missing translations or unsourced statistics
  2. Procedures are checked against the product before they are written down
  3. The founding team owns the checks and edits what needs editing
  4. Our founders take full responsibility for published content

Our human editorial team

AI used for our content

Writing model provider
Anthropic
Model
Claude
Primary use
Content creation and research (translation runs on other models)
Model training
Not used for model training
Video generation
Terms: AI providers and media handling

Our AI Authors

Meet the AI personas behind our content.

Each AI author has its own subject area and writing style.

Meet our team

Questions about how we use AI

Is the content on this site written by AI?

The drafts are. Tutorials, guides and blog posts are written with AI under one of three named personas, then put through automated style, source and translation checks that refuse the ones that fail before anything is committed.

Which model do you use, and does it train on our data?

We draft with Claude. Translation runs through a different model: the pipeline calls OpenAI for the site copy and Gemini for some app content, so the model that writes and the model that translates are not the same one. In each case the data we send is not used to train the model, which is the condition under which we use it.

What is an AI persona, and is it pretending to be a person?

A persona is a fixed subject area and writing register attached to a byline. Damien, Alexis and Henry are labelled as AI authors on their own pages, and they do not claim jobs, employers or personal experiences.

How are the translations made?

AI translates the site into its other languages from the English source. That is how 18 locales stay in step, and it is also why a translation can trail the English version by a short period after an edit.

What are the limits of this approach?

AI drafting is good at structure, coverage and translation, and it is unreliable about specifics it was not given. So we hold it to sourced claims and tested procedures, and where a first-hand measurement does not exist we leave the number out rather than let the model supply one.

What this means for you

You can treat our procedures and product facts as checked by a person who is accountable for them, and you can see which author drafted any page. What you should not treat this content as is personal testimony or independent journalism, because it is neither. If a claim on this site matters to a decision you are making, write to us and we will point you at the source it came from.

Contact

We welcome questions about our AI practices. Write to us.