Home Blog Monetisation

YouTube’s AI disclosure rules for news channels

Most AI use on a news channel does not need a label. The few cases that do are exactly the ones a news channel is most likely to stumble into.

News channels use AI everywhere now: drafting scripts, writing titles, generating thumbnails, narrating with synthetic voices. Most of that does not need to be disclosed. A small set of uses does — and they are precisely the uses that look like reporting.

This guide follows YouTube’s disclosure guidance for altered or synthetic content. Read that page for the current wording before you rely on any detail.

When disclosure is required

YouTube asks creators to disclose content that is realistic and made or meaningfully altered with AI or other tools, where a viewer could mistake it for real. Its examples include content that:

  • makes a real person appear to say or do something they did not do;
  • alters footage of a real event or place;
  • generates a realistic-looking scene that did not actually happen.

Read that list as a news editor and every item is a news risk. An AI-generated clip of a politician “speaking”, a synthetic image of a flood that looks like the real one, a fabricated scene of a protest — these are exactly the things a viewer would take as evidence.

When it is not required

YouTube lists uses that do not need a label, including clearly unrealistic content, minor edits such as colour correction, and using AI for productivity — scripts, ideas, captions. Cloning your own voice to narrate your own script is also listed as not requiring disclosure.

So for a typical faceless news channel: an AI-assisted script, AI-suggested titles and an AI narration voice reading your own words are not in the categories YouTube lists as needing a label. Footage or images that realistically depict a real event or real person doing something they did not do are.

The line is not “did you use AI”. It is “could a viewer mistake this for a real recording of something that happened”.

Why news has to be stricter than the rule

The disclosure rule is a floor. A news channel’s credibility depends on viewers trusting that what you show them is real, and the consequences of a realistic fake are worse on news than anywhere else: it can be clipped, shared and cited as evidence by people who never saw your label.

  • Never generate realistic footage of a real event. If you need a visual, use real footage you have rights to, a map, a timeline or an illustration that is obviously an illustration.
  • Never put words in a real person’s mouth. Not as a joke, not as a “reconstruction”. Quote them on screen as text instead.
  • Keep AI thumbnails honest. A generated image that shows an event as having happened when it has not is misleading whether or not it is labelled — see thumbnails for news videos.
  • Label illustrations as illustrations. If you use a generated image for atmosphere, say so on screen.

How to disclose

Disclosure is a setting in YouTube Studio during upload. According to YouTube’s guidance, the label appears on the video player itself for photorealistic content, and in the expanded description otherwise. YouTube says disclosure does not affect a video’s eligibility for monetisation.

Disclosure is a separate question from the monetisation policies. An AI-narrated channel with original scripts can be both undisclosed and monetisable; a channel that mass-produces identical AI videos can be fully disclosed and still fail the inauthentic content policy.

A checklist before you publish

  1. Does any image or clip realistically show a real event that I did not film or license? Replace it, or disclose it and label it on screen.
  2. Does anything show a real person saying or doing something they did not? Remove it.
  3. Is the AI use limited to scripting, titles, captions or narration of my own words? No label needed under YouTube’s listed examples.
  4. Would a viewer who clipped ten seconds of this video mislead someone else with it? If yes, change the video.

For the production side of an AI-assisted news channel, see the faceless news channel workflow. VidiScoop uses AI for drafting scripts, packaging and thumbnail images, and fact-checks every script claim against the reporting it came from — but what you publish, and how you label it, stays your call.

Frequently asked

Do I need to disclose an AI voice on my news channel?

YouTube lists cloning your own voice as not needing disclosure, and narration of your own script is a productivity use. Making a real person appear to say something they did not is what requires disclosure — and on a news channel, should not be done at all.

Do AI-written scripts need a label?

No. YouTube lists using AI for scripts, ideas and captions as not requiring disclosure. The script still has to be accurate and add value under the monetisation policies.

Does disclosing AI content hurt monetisation?

YouTube says disclosure does not affect monetisation eligibility. Whether the content itself is original enough is a separate review.

Monetisation

Copyright, fair use and news footage on YouTube

News reporting is one of the purposes fair use is meant to protect — and news channels still collect more copyright claims than almost anyone. Here is why, and how to stay out of trouble.

Monetisation

Reused content and news channels: what YouTube actually allows

“Reused content” is the most common reason news channels are refused monetisation. The policy is narrower than its reputation — here is what it targets and what it does not.

Built on exactly this method

Let the research run itself.

VidiScoop does the sourcing, deduplication and scoring described here, then writes the packaging — so your time goes to the decision and the video.

No credit card required · Your first ranked scan takes about a minute