15 essential tasks GPTs can do for journalists

This article was migrated from an old version of our website in 2025. As a result, it might have some low-quality images or non-functioning links - if there's any issues you'd like to see fixed, get in touch with us at info@journalism.co.uk.

Generative AI tools such as GPTs are becoming increasingly smart, but their use still requires careful oversight. These models can accelerate routine tasks, freeing journalists to focus on impactful reporting. However, always fact-check outputs, as GPTs can generate plausible but inaccurate or biased information.

Key tasks GPTs can perform for newsrooms

Summarising large texts: Quickly condense lengthy reports, studies, or documents into key points or quotes. Useful for rapid research and briefing.

• Read more: How AI can help journalists track MPs financial interests

Generating questions and research prompts: Suggest interview questions, research angles, or expert names. Ideal for unfamiliar topics, but verify all suggestions for accuracy as there have been reports of "expert" names that were totally made up

Drafting headlines and content optimisation: Produce headline variations for A/B testing or brainstorming. GPTs can also generate SEO titles, URLs, and notification summaries to improve discoverability. Always double-check for clarity and accuracy.

Translating content: Provide rough translations to understand foreign-language material. For publication, use professional translation or specialist tools.

• Read more: How a Greenlandic publisher uses its own AI translator to boost subscriptions

Creating social media posts: Generate posts for platforms such as X (formerly Twitter) or LinkedIn, saving time on repetitive content creation.

Providing context and explainers: Offer background information or plain-language explainers for complex topics. Always cross-check facts and sources.

Assisting with data analysis and visualisation: Help interpret datasets, suggest story angles from data, or draft simple charts and visualisations. Outputs should be reviewed by data specialists.

• Read more: What can generative AI actually do for ordinary journalists?

Drafting emails and subject lines: Speed up routine communications with draft emails or subject lines. Edit for tone and accuracy before sending.

Additional emerging use cases

Personalised news recommendations: Curate news feeds and alerts tailored to individual user interests, increasing engagement and retention.

Multimodal storytelling: Generate captions, descriptions, and narratives for images, videos, and multimedia, supporting richer, more interactive storytelling.

• Read more: What you need to know about modular journalism

Real-time fact-checking: Assist in verifying claims, statements, and social media posts to identify misinformation and ensure accuracy in reporting.

De-jargonising complex information: Translate technical or scientific content into accessible language for broader audiences.

Augmented investigative reporting: Provide research assistance, background context, and help analyse public records for investigative journalism.

• Read more: How the BBC is using artificial intelligence

Automated routine news updates: Draft basic news briefs from structured data (e.g., weather, sport, financials), freeing up journalists for deeper reporting.

Ethical standards and transparency: Support the development of newsroom guidelines for responsible AI use, including disclosure and human oversight.

Limitations and best practices

GPTs struggle to understand context or intent; they generate text based on patterns in data.

Always verify facts, quotes, and names. Be alert to potential bias or fabricated information.

Use GPTs as assistants, not as replacements for journalistic judgement or original reporting.

This article was first published on 31 January 2023. It was updated on 18 September 2025 with the help of NOAN, an AI assistant, before it was edited by a human