Building the newsroom AI playbook with journalists in the loop

AI in journalism is often discussed in extremes, either as a revolution that will transform reporting or a threat to the profession.

In this episode of The Media Copilot podcast, Pete Pachal speaks with Gina Chua, Executive Editor at Large at Semafor, where she leads AI strategy, and Executive Director of the Tow Center for Digital Journalism at CUNY. The conversation focuses less on theory and more on how AI tools are being used in everyday newsroom workflows.

Listen or watch:

• YouTube
• Spotify
• Apple Podcasts

At Semafor, some of the systems Chua describes are simple tools designed to reduce time spent on routine tasks. These include internal tools for copy editing and proofreading, and a workflow that reads a story draft and suggests datasets that could help illustrate it. The reporter can then decide whether to explore those datasets or build a chart using existing newsroom tools.

Another example involves helping reporters find related coverage more quickly. A tool reads a draft story and searches across publications for related reporting, sometimes across languages. It then returns summaries of potential articles so editors can decide which ones are worth reviewing.

Semafor also works with large volumes of interview transcripts from events and panel discussions. Chua explains how those transcripts are integrated into Slack so reporters can search them directly. Journalists can ask whether a speaker mentioned a particular topic, locate the section of the transcript where it appears, and review it themselves. The next step the team is experimenting with involves automated alerts that notify reporters when certain topics appear in interviews across the archive.

Elevating the human element

Throughout the conversation, Chua emphasizes that these tools are meant to assist newsroom work rather than replace editorial judgment. When discussing publishing standards, she says the key principle is that anything published must be reviewed by a human editor who takes responsibility for it.

She also offers a practical way to think about large language models. They are designed to handle language, not verify facts. When used with a defined body of text, such as a transcript or speech, they can be helpful for tasks like summarizing, comparing passages, or identifying themes. Problems are more likely to appear when they are asked to generate information outside of that context.

The discussion also turns to how audiences are already encountering AI in everyday information environments. Chua points to research showing a small but growing number of younger users say they get news from chatbots. She argues that the shift may be larger than the statistics suggest, since AI summaries are now embedded in search results and other tools people use to find information.

For news organizations, that raises broader questions about how people will access information in the future. If AI systems increasingly sit between audiences and publishers, journalism will need to think carefully about how reporting reaches readers and how trust is maintained.

Why this matters

AI tools are already influencing how newsrooms work and how audiences discover information. Understanding the practical uses and limitations of these systems is becoming part of the daily reality of journalism.

• Subscribe to our newsletter

  How AI is changing media, journalism, and content creation.

  Learn More

What we cover

• How Semafor is experimenting with AI tools inside the newsroom
• Using transcripts and Slack to search interviews and discussions
• Why language models are useful for handling text but not verifying facts
• The role of human review in newsroom publishing decisions
• How AI interfaces are changing the way audiences find news

Key takeaways

• AI tools in newsrooms are often used for efficiency rather than automation
• Large language models work best when applied to known text sources
• Human oversight remains central to publishing decisions
• AI driven discovery tools may increasingly shape how audiences access journalism

About the 👤 Guest: Gina Chua

LinkedIn 👉 https://www.linkedin.com/in/ginachua

X (Twitter) 👉 https://x.com/GinaSKChua

Instagram 👉 https://www.instagram.com/gina_chua_nyc 

Personal Website / Writing 👉 https://ginachua.me 

Author Page (Semafor) 👉 https://www.semafor.com/author/gina-chua

📰 Semafor Official Website:  https://www.semafor.com

About the show: To explore more conversations like this and see what’s new, visit the freshly updated Media Copilot website at mediacopilot.ai. You’ll find new episodes, expanded resources, and tools designed for journalists, communicators, and media leaders navigating the fast-changing world of AI. It’s the home base for everything Media Copilot and it’s just getting started.

Enjoyed this episode?

Subscribe to The Media Copilot on Substack, Apple Podcasts, Spotify, or your favorite app. On YouTube?  Tap the Like button and Subscribe to the YouTube channel.

For more AI tools and resources built for media professionals, visit MediaCopilot.ai.

Produced by Pete Pachal and Executive Producer Michele Musso
Edited by the Musso Media Team 

Music: “Favorite” by Alexander Nakarada, licensed under CC BY 4.0

All rights reserved. © AnyWho Media 2026