Editorial and AI-use policy

Articles on Two-Minute Speaking are drafted with the help of AI tools, including Claude, inside a pipeline the founder controls end to end. AI never publishes on its own: every draft is fact-checked against its cited sources, reviewed against an editorial rubric, and then read, edited and approved by a person — Ali Saberian — before it goes live. This page explains exactly how that works and how to report a mistake.

Do you use AI to write articles on this site?

Yes. Drafts are generated with AI language models, primarily Claude, based on research and source material gathered for each topic. AI speeds up the first draft; it does not decide what gets published. Every article that reaches this site has been through the review steps below, and every one is read and edited by a human before it appears.

How do you check facts before publishing?

Every statistic, quote and specific claim in a draft is checked against the source it cites. A script re-fetches each cited source directly and compares the claim in the draft against the live page, flagging anything that does not match or that can no longer be verified. Drafts are also scored against an internal editorial rubric that checks structure, tone and whether claims are properly sourced before a human ever sees them.

Who has final approval before something goes live?

Ali Saberian, the founder, reads, edits and approves every article before it is published — nothing on this site goes live automatically. Statistics and quotes always link to their primary source so you can check them yourself. Product claims are limited to what Two-Minute Speaking actually does today; we do not publish claims about results, scores or outcomes the product has not measured.

Do you run sponsored or affiliate content?

No. [OWNER: confirm] Two-Minute Speaking does not currently publish sponsored posts, paid placements or affiliate links on this site. If that changes, any paid or affiliate content will be labelled clearly at the top of the article, not mixed in with regular editorial content.

Are the images on this site AI-generated?

Some are. Illustrations on this site may be generated with AI tools rather than photographed or drawn by hand. Where that is the case, it is noted in the image’s own metadata. We do not use AI to alter real photographs of people or to fabricate images presented as documentary evidence.

What happens if an article has an error?

Email privacy@aienglishspeaking.app with the article title and what looks wrong. We check it against the source and correct it if it is wrong. Corrections are noted at the end of the article itself, with the date the correction was made, so the history of a change stays visible rather than silently disappearing.

Spotted a mistake? Email privacy@aienglishspeaking.app and tell us which article and what looks wrong — we will check it and, if needed, correct it.