AI for Newsrooms
Real-world AI projects from newsrooms worldwide, curated news about AI in newsrooms, case studies, reports, researches and guides to help journalists and news organizations understand and implement AI.
AI for Newsrooms. News, projects, reports, guides:
It’s bots vs. reporters at the AP
The Associated Press (AP) is debating the use of artificial intelligence (AI) in journalism, with some leaders believing it is inevitable and others resisting the change. Aimee Rinehart, AP's Senior Product Manager for AI, suggested that AI could generate stories, saving reporters time on writing. However, some journalists are alarmed by the idea, arguing that strong reporting and writing are essential to journalism. The debate reflects a broader conflict in the media industry over how AI should be applied, with some companies embracing it and others wary of its potential threat to jobs.
Semafor
Radar Antigênero
Radar Antigênero is a free AI-powered monitoring platform that tracks and analyzes hate and “antigender” speech on YouTube targeting women, girls and LGBTQ+ people. Developed by newsroom Gênero e Número with Novelo Data, it continuously collects videos using hate-related keyword searches, auto-transcribes them with OpenAI’s Whisper, and stores the transcripts in a database for manual and AI-assisted classification. Experts in technology, data science and gender studies helped refine its methodology so users can search videos from 2018–2026, see channels that systematically spread anti-gender discourse, and explore how narratives spread through views, likes and producers. Organized by thematic axes, discursive strategies and main targets, Radar provides evidence and context for research, journalism and public policy debates on gendered hate in Brazil.
journalism🇧🇷 · 6 initiatives
Press releases to news
ANKA News Agency built a simple but powerful AI-assisted workflow that automatically transforms routine municipal press releases into publishable news drafts in the agency’s house style, using carefully engineered persistent prompts in ChatGPT’s Projects feature instead of complex fine-tuned models. This setup lets reporters feed bulletins into a standardized prompt that remembers ANKA’s context and style, producing fast, consistent drafts that journalists can then refine, dramatically reducing “robot work” and freeing up time for exclusive reporting and deeper analysis. Along the way, the project also acted as a change-management tool, normalizing AI use in the newsroom by showing skeptical colleagues that well-prompted automation can improve speed and quality without replacing journalists, and giving participants a broader, more strategic understanding of where automation fits into the future of journalism.
content production🇹🇷 · 1 initiative
Newsroom Policies for AI in Journalism
The third briefing from the AI and Journalism Research Working Group finds that organizational AI policies tend to prioritize principles and values over practical guidance.
CNTI
Legitalk
Legitalk is CT Mirror’s custom AI-powered meeting coverage tool that automatically ingests videos of Connecticut legislative hearings from YouTube, transcribes them with Deepgram, and uses OpenAI to create bill-focused, timestamped summaries so reporters can instantly jump to the exact moments they need. Instead of scrubbing through eight– to ten-hour meetings, journalists search by bill number, keyword, or speaker in a three-panel interface (bill summaries, full transcript, video player) and play from any highlighted section, turning a full-day listening task into a couple of hours of targeted reading and verification. Built by data reporter Angela Eichhorst on top of the open-source Golden Dome project and tailored to CT Mirror’s committees and workflows, Legitalk doesn’t replace reporting; it surfaces the “needle in the haystack” so humans can do the interviewing, fact-checking, and storytelling faster and at greater scale.
distribution🇺🇸 · 48 initiatives
What does RAG mean for news?
Retrieval-Augmented Generation — or RAG — is changing how information is surfaced online. By pairing large language models with external data sources, it turns AI systems from static text predictors into tools that consult documents and databases before responding. For news publishers, that shift could redefine what it means to be discoverable.
Tomorrow's Publisher
Weekly Newsletter.
One Friday email that rounds up AI for Newsroom.
Chat with AI For Newsroom
Let's Build the AI Index for Newsrooms
Submit AI initiatives your newsroom is working on, resources, policies, or media content to help journalists and newsrooms worldwide stay ahead in the AI era.