Artificial intelligence is increasingly being used across the media and publishing industry. The Nation Media recognises that AI tools can assist journalists, editors and other members of our organisation with certain tasks, but they also present significant risks to accuracy, originality, privacy, transparency and public trust.
This AI Policy establishes the principles that govern the use of artificial intelligence in connection with journalism and other content published by The Nation Media.
Our objective is to use AI where it provides legitimate editorial or operational value while ensuring that human judgement, accountability and verification remain at the centre of our publishing process.
1. Human Editorial Responsibility
AI does not replace editorial responsibility.
A journalist, editor or other appropriately authorised member of The Nation Media remains responsible for content that is published under our name, regardless of whether AI tools were used during its preparation.
AI-generated or AI-assisted material must therefore be reviewed by a human before publication where it forms part of editorial output.
The use of an AI system does not transfer responsibility for accuracy or editorial decisions to the technology provider.
2. Permitted Uses of AI
Subject to appropriate human review, AI tools may be used to assist with activities such as:
- Brainstorming story ideas or questions;
- Organising notes and information;
- Summarising material for internal editorial use;
- Transcribing recordings where appropriate;
- Translating or assisting with translation;
- Improving grammar, spelling and readability;
- Generating administrative or routine content;
- Assisting with research workflows;
- Analysing large amounts of non-sensitive information;
- Creating internal drafts or outlines; and
- Supporting technical, production or accessibility tasks.
The use of AI must always be proportionate to the task and consistent with editorial standards.
3. AI-Generated Text
AI may produce text that appears convincing while containing inaccurate, incomplete or entirely fabricated information.
For this reason, AI-generated text must not be treated as an authoritative source.
Journalists and editors should independently verify important factual claims, names, dates, quotations, statistics, locations, legal matters and other material information before publication.
AI output should be treated as a tool or draft rather than as verified journalism.
4. Fact-Checking AI Output
Particular care should be taken when an AI system provides:
- News claims;
- Historical information;
- Statistics;
- Scientific or medical information;
- Legal information;
- Financial information;
- Quotes attributed to individuals;
- References or citations;
- Names and identities;
- Links or source references; or
- Information about developing events.
AI systems can produce plausible but false information, sometimes referred to as hallucinations.
Any material fact that could materially affect a published report should therefore be checked against reliable sources before publication.
5. Fabricated Quotes and Sources
The Nation Media must not knowingly publish an AI-generated quotation as though it were spoken or written by a real person.
AI tools must not be used to invent:
- Interviews;
- Sources;
- Eyewitness accounts;
- Statements;
- Citations;
- Documents;
- Events;
- Statistics; or
- Other evidence presented as genuine.
A generated example, hypothetical statement or fictional scenario must never be presented to readers as an authentic news source.
6. AI and News Reporting
AI-generated material must not be used as a substitute for reporting.
Where a story requires interviews, source verification, on-the-ground reporting, document review or independent confirmation, those journalistic functions remain the responsibility of the relevant editorial team.
AI may support a journalist’s workflow, but it should not create the appearance that reporting has taken place when it has not.
7. AI-Generated Images and Video
AI-generated or materially AI-manipulated images and video require particular care in a news environment.
The Nation Media should not present synthetic or substantially altered visual material as genuine documentary photography or authentic news footage.
Where AI-generated or materially AI-altered visual content is published for legitimate editorial purposes, readers should be given appropriate context so that they are not misled about its origin or authenticity.
AI may be used for clearly illustrative purposes where the presentation makes the illustrative nature of the material reasonably apparent.
8. Deepfakes and Synthetic Media
We recognise that AI can create highly realistic synthetic photographs, audio recordings and videos.
Such material should be treated with caution, particularly when it appears to show a public figure, political leader, organisation or other identifiable person making statements or taking actions.
Before publishing potentially consequential synthetic media, reasonable steps should be taken to establish its authenticity, provenance and context.
Material that cannot be adequately verified should not be presented as authentic simply because it appears convincing.
9. AI-Assisted Photography and Editing
Ordinary technical adjustments to photographs, such as cropping, resizing, exposure correction or similar production changes, should be distinguished from generative alterations that add, remove or substantially change elements of an image.
Generative editing should not be used to create a false impression of a real-world event.
Where significant AI manipulation is editorially relevant, appropriate disclosure or contextualisation should be considered.
10. Disclosure and Transparency
Not every internal use of AI requires a public disclosure.
For example, routine spelling assistance, transcription or internal organisation may not require a notice to readers.
However, transparency should be considered where AI has made a meaningful contribution to the published material or where failing to disclose its use could reasonably mislead the audience.
Where appropriate, The Nation Media may use wording such as:
“This material was produced with the assistance of artificial intelligence and reviewed by our editorial team.”
The form and level of disclosure may vary according to the nature of the content.
11. AI and Editorial Independence
AI tools must not determine the editorial position of The Nation Media.
Editors and journalists remain responsible for decisions concerning:
- What is newsworthy;
- Which sources are credible;
- What information should be published;
- How a story should be framed;
- Which claims require further verification; and
- Whether publication is in the public interest.
AI recommendations should not be treated as neutral or inherently authoritative.
12. Bias and Fairness
AI systems can reproduce or amplify biases present in their training data, design or underlying models.
Where AI is used to assist with editorial work, users should remain alert to potential bias involving individuals, communities, countries, cultures, political views, social groups or other characteristics.
AI output should not be accepted simply because it is fluent, confident or professionally written.
Editorial judgement should be used to identify unsupported assumptions, stereotyping and potentially discriminatory conclusions.
13. Confidential and Sensitive Information
Journalists, editors, contributors and other staff should exercise particular caution before entering information into external AI systems.
Confidential sources, unpublished investigations, private communications, personal information, commercially sensitive material, credentials and other protected information should not be entered into an AI service unless there is an appropriate and authorised basis for doing so.
Where an AI provider may retain, analyse or use submitted information for model improvement or other purposes, this should be considered before any non-public material is submitted.
14. Personal Data
AI tools should not be used to process personal information unnecessarily.
Where personal information is involved, appropriate safeguards should be considered, including whether the information can be removed, anonymised or minimised before being submitted to an AI system.
The Nation Media’s Privacy Policy provides additional information concerning the handling of personal information.
15. Copyright and Intellectual Property
AI tools do not automatically provide permission to reproduce copyrighted material.
Staff and contributors should not use AI systems to circumvent copyright restrictions or reproduce protected material without an appropriate legal or editorial basis.
Care should also be taken when using AI-generated material because questions concerning copyright ownership and originality may vary according to the circumstances and applicable law.
Where third-party material is used in connection with AI-assisted content, appropriate rights, permissions and attribution should be considered.
16. AI Translation
AI-assisted translation may be used where appropriate, but important editorial material should be reviewed by a suitably competent person before publication where accuracy is significant.
Particular care should be taken with:
- Names;
- Legal terminology;
- Quotes;
- Cultural references;
- Idioms;
- Political terminology;
- Technical information; and
- Statements where a small translation error could materially change the meaning.
17. AI Transcription
AI transcription tools may be used to convert interviews, press conferences, speeches or recordings into text.
Transcripts should be checked before quotations are published.
An AI transcription error must not become a published quotation or materially alter what a person actually said.
Where audio quality is poor or a statement is unclear, the original recording should be consulted.
18. AI and Research
AI may assist with preliminary research, but an AI-generated answer is not itself evidence that a claim is true.
Journalists should consult and, where appropriate, cite the underlying primary or authoritative sources rather than relying solely on an AI-generated summary.
AI-generated references should be independently checked before being used in published journalism.
19. Automated Content
Where automated systems are used to generate or update routine information, appropriate safeguards should be established to reduce the risk of publishing incorrect information.
Automated publishing should be subject to suitable monitoring, particularly for information that can change rapidly.
Materially important editorial decisions should remain subject to human oversight.
20. Corrections Involving AI
If AI-assisted or AI-generated material contributes to a published error, The Nation Media may correct the material in accordance with its editorial corrections procedures.
Where an AI-related error is significant, we may also review the process that led to the error and introduce additional safeguards to reduce the likelihood of recurrence.
The fact that AI was involved does not reduce our responsibility to address a material publishing error.
21. Staff and Contributor Responsibilities
Anyone producing content for The Nation Media is expected to use AI tools responsibly.
Staff and contributors should:
- Verify significant factual claims;
- Avoid fabricated material;
- Protect confidential information;
- Respect copyright and intellectual property;
- Consider potential bias;
- Follow editorial standards;
- Maintain appropriate human oversight; and
- Be transparent about significant AI involvement where appropriate.
Individuals may be asked to disclose their use of AI where this is necessary for editorial review or compliance purposes.
22. AI Detection
AI-detection software may be used as one potential indicator when reviewing content, but it should not be treated as conclusive proof that a particular piece of writing was or was not generated by AI.
AI detectors can produce both false positives and false negatives.
Editorial decisions should therefore rely on evidence, context and human review rather than an automated detection score alone.
23. Use of AI by Readers and Contributors
Readers and contributors may use AI tools when preparing material for submission, subject to the requirements applicable to user and contributor content.
Submitting AI-generated material does not remove the contributor’s responsibility for its accuracy, originality, legality or compliance with our publishing requirements.
Contributors must not submit AI-generated material containing fabricated sources, invented quotations or deliberately misleading information.
24. AI-Generated Comments and Spam
Automated systems should not be used to flood comment sections, contact forms or other interactive areas with repetitive or deceptive material.
AI-generated comments, reviews, messages or submissions that are designed to manipulate discussion, impersonate genuine users or distribute spam may be removed.
Accounts or users repeatedly engaging in such behaviour may have access to relevant features restricted.
25. Use of AI in Advertising and Commercial Content
AI may be used in the preparation of advertising, marketing or commercial material where permitted.
Advertisers and commercial partners remain responsible for the accuracy and legality of claims made in their advertising.
AI-generated promotional material must not be used to create deliberately deceptive claims, fabricated endorsements or misleading representations.
Commercial material must remain distinguishable from independent editorial journalism.
26. Emerging AI Technologies
Artificial intelligence technology continues to develop rapidly.
New tools may introduce capabilities and risks that are not specifically addressed in this policy.
The Nation Media may therefore establish additional internal guidance or restrictions when new AI technologies, applications or risks arise.
27. Our Commitment
The Nation Media believes that technology should support journalism rather than undermine public trust in it.
We will seek to use AI thoughtfully, with human oversight, appropriate verification and consideration for accuracy, transparency, privacy, originality and editorial independence.
Our approach may evolve as technology, industry practice and applicable legal requirements develop.
28. Policy Updates
This AI Policy may be reviewed and amended periodically to reflect changes in technology, editorial practice, legal requirements and our use of AI tools.
The current version will be published on this page.
Last Updated: 12 September 2026