Frequently asked questions

It is not an authoring tool you have to learn, and it does not write your teaching for you: you decide what is in the lesson, and whether it is good enough.

A lesson is built around one core learning job per page: a Key idea and a Takeaway where the explanation lands, predictions, checks and practice, and diagrams or tables only where the material genuinely has a structure, comparison, process or prediction worth showing. Worthwhile but non-core material goes into Go deeper, an optional section at the end.

PowerPoint (.pptx), Word (.docx) and PDF (.pdf) files up to 300 MB, and Markdown or plain text files up to 25 MB. Images, video and embedded files are left out. A PDF is read as printed: any notes printed on its pages are read too, so for speaker-notes control upload the PowerPoint. One lesson carries up to 120 extracted sections and about 180,000 characters. Above that, LessonLayer asks you to split the material rather than silently dropping part of it.

You can also build a whole module at once on the module page: add each lecture (a lecture can be several files), and LessonLayer plans the module in one go, matching ready-made interactive activities to lectures where one fits so that lessons do not repeat one another by accident. You review and edit that plan, then each lesson is made, reviewed and downloaded exactly like a single lesson, and charged the same way.

Nothing for your students downloads until you have been through the review screen. There you can reorder, promote, demote or remove any section. Anything the academic checks flagged must be confirmed first, and a few items cannot be confirmed away at all: those sections have to go, or the lesson is generated again. A sentence LessonLayer could not have checked (for example while its checker was unavailable) needs its own decision. If you keep a flagged item, the downloaded file is marked "lecturer-approved" rather than "verified". Material the lesson does not use is listed under "Not used by the generated lesson", so the omission stays visible and yours to overrule.

You can change any text. The change is made in your browser, saved in the file you download and marked "edited by you". Edited text is yours: it is not checked again by LessonLayer, and the file is marked "lecturer-approved". When you download, your edits travel with the lesson to our service to be checked (and charged, if the lesson has not already been charged), and are not kept. The underlying lesson file can be downloaded at any time.

The examples, questions and practice cases that make a lesson interactive are written by LessonLayer: every sentence LessonLayer writes goes through an independent check before download, and anything still needing your judgement is flagged, not quietly accepted. You review all of it before your students see it.

The HTML file is standalone, with its styles and interactions inside it: no plugin, no login, no account for your students. You share it the way your department already shares teaching material. The underlying lesson file is downloadable too. Ready-made activities are built into the file and work offline.

The SCORM 1.2 package reports to your Virtual Learning Environment, and only to it, whether the student has completed it (seen every page of the core route; the optional Go deeper pages do not count) and the lesson's own place record, so that a student can carry on where they left off. There is no LTI integration and no institutional single sign-on.

"Completed" means the student has visited every page of the core route. It does not mean they answered every activity: answers and scores are not checked for completion and are never sent to your Virtual Learning Environment.

PowerPoint and Word files are unpacked locally, and a PDF is read locally too. Only the extracted text, the file name and type, any outcomes you typed and your key idea, if you gave one, go over HTTPS to our generation service, and on to the AI provider for your department: Anthropic, which runs the Claude model that writes the lesson, in the US. Anthropic keeps what it receives for at most 30 days (longer only if it is flagged for misuse), and does not train its models on it. Your speaker notes go too only if you choose to include them.

While your lesson is being made, the service keeps the job in Cloudflare storage in the EU, so closing the tab does not lose it; it is deleted when you download the lesson, or 24 hours after you started it at the latest, and it is never used for anything else. See the privacy page, written against what the code does.

Access is provisioned manually, not through self-service sign-up. See the Beta pilot page.

Lessons are paid from your department's prepaid AI balance, at the AI provider's cost with no markup. The Make a lesson page shows an estimate before you start and the actual cost when the lesson is ready. A lesson is charged when it is built and ready to use, whether or not you download it, so a lesson that fails or is blocked costs nothing. A lesson that needs your decision first is charged when you first download it. The balance does not expire while your department has access, and any unused balance is refunded on request. See the terms.

There are no student accounts. No learner personal data and no cross-session tracking. A SCORM package reports completion and the student's place in the lesson only to your own Virtual Learning Environment.