Guidelines by field
AI guidelines in journalism and media what the field's own bodies ask
Newsrooms wrote some of the earliest AI policies, because their whole value rests on readers trusting that what they publish is real and checked. The rules are almost entirely voluntary, and remarkably consistent.
Last reviewed 27 October 2026. Every claim links to its source.
At a glance
AI output is unvetted source material
It gets the same checking as a tip from a stranger.
Humans stay in editorial control
Outlets remain responsible for everything they publish.
Tell audiences when AI use is significant
And keep a public record of the systems used.
Never fake reality
Do not add or remove elements from news photos, video or audio, and label AI illustrations clearly.
Pay for journalism used by AI
AI companies should use journalism only under agreements that credit, respect copyright and pay fairly.
Who has spoken
- Binding law, regulation, court ruling or enforceable professional rule
- Voluntary guidance, principles or a code an organization chooses to follow
Voluntary Nov 2023
Reporters Without Borders and 16 partners
Paris Charter on AI and Journalism
Ten principles: human editorial control, accountability, disclosure, provenance checks, no fabricated reality, and fair pay for journalism used by AI.
Voluntary Aug 2023
Associated Press
Standards on generative AI
AI output is treated as unvetted source material; no altering of news images, video or audio; AI illustrations labeled.
Voluntary May 2023
Reuters
Guidance to journalists
Discloses when AI use is material, and bars pasting unpublished stories or confidential source material into public AI tools.
Voluntary Oct 2023; Feb 2024
BBC
Generative AI principles and editorial guidance
Three principles and editorial guidance centered on the public interest, transparency and human oversight.
Voluntary Autumn 2026
Society of Professional Journalists
Revised Code of Ethics
The first revision since 2014, endorsed by the board in September and approved by 93% of voting members, with the result announced at the October convention. Journalists are responsible regardless of the tools, without naming specific technologies so the code stays current.
Speed, technology or medium do not excuse inaccuracy or unfairness.
Binding and voluntary
Binding
Law enters only at the edges: copyright, defamation and the EU AI Act's transparency duties for deepfakes. Newsroom policies bind staff as employment rules.
Voluntary
The Paris Charter, the SPJ code and newsroom standards are voluntary professional commitments.
What changed in 2025 and 2026
- A study by 22 public broadcasters in 18 countries found 45% of AI assistants' answers about news had at least one significant issue, which strengthened newsrooms' insistence on verification.
- The SPJ adopted its first code revision in twelve years in autumn 2026; members' approval was announced in October.
Questions to ask yourself
- Did a human check every fact, quote and attribution in this piece against a primary source?
- Would a reader feel misled if they learned how AI was used here?
- Could this image or audio be mistaken for a real recording?
How this lines up with my everyday guidelines
| My everyday guidelines | The field's nearest rule |
|---|---|
1. Maintain your expertise Always review and verify outputs, especially for client-facing work. | Treat AI output as unvetted source material (AP; Reuters). |
2. Practice transparent collaboration Do not present AI-generated content you would not confidently stand behind. | Outlets remain responsible for everything they publish (Paris Charter; SPJ). |
4. Use for good Never use AI to deceive, manipulate or harm others. | Never add or remove elements from news images, video or audio (AP). |