🧭 AP Just Updated Its AI Rules. Here’s a Useful Test for Your Newsroom.
The Associated Press has updated its newsroom standards for AI — and the interesting part isn’t that AP is using AI.
It’s where it draws the line.
AP now explicitly allows AI to assist with tasks such as:
- early-stage research and document summarisation
- transcription and translation
- suggesting headlines, summaries and shotlists
- grammar, spelling and search optimisation
But the output must still be reviewed and edited by journalists.
And AP remains clear that AI does not replace reporting, sourcing, verification or editorial judgement.
Instead of asking “Can we use AI for this?” ask these five questions:
1️⃣ Is AI doing the journalism — or supporting the journalism?
Using AI to organise 200 pages of documents is very different from asking it to decide what the story is.
2️⃣ Is a journalist still accountable for the result?
If nobody can clearly say, “I checked this and I’m responsible for publishing it,” the workflow probably needs redesigning.
3️⃣ Can the output be independently verified?
A transcript can be checked against audio.
A summary can be checked against the source document.
An invented “expert quote” cannot.
4️⃣ Does AI materially change what the audience sees?
AP’s updated rules include disclosure standards where generative AI plays a material role in published content.
Using AI privately to clean up notes is not the same thing as using synthetic material in published journalism.
5️⃣ Are some things simply off-limits?
AP continues to prohibit generative AI from creating, altering or enhancing news photographs.
Every newsroom should identify its own red lines rather than waiting for problems to happen.
🛠️ Try this with your team
Take one recurring newsroom task — transcription, headline writing, research, social copy, translation or data analysis — and paste this into your AI tool:
I am designing an AI policy for the following newsroom task:
[TASK]
Assess the workflow using these questions:
1. What should AI be allowed to do?
2. What must remain a human responsibility?
3. What must be independently verified?
4. When should AI use be disclosed to the audience?
5. What uses should be prohibited entirely?
Then suggest a short policy rule we could realistically apply in a newsroom.
Do not assume that efficiency outweighs accuracy, ethics or editorial accountability.The goal of an AI policy shouldn’t be to produce a giant document nobody reads.
It should make the right behaviour obvious when a journalist is under deadline pressure.
That’s where AP’s updated approach is useful: AI can assist with more newsroom work, while responsibility stays firmly with the journalist.
🔗 AP’s updated newsroom AI standards
Read AP’s updated newsroom AI standards
🚀 Social Weaver helps newsrooms develop practical AI policies, workflows and training that journalists can actually use.
Explore Social Weaver services
💬 Prefer WhatsApp? Message us directly: