New York Times Uses AI to Monitor Podcasts with ‘Manosphere Report’ Tool
The New York Times revealed on February 11, 2026 that it is using an internal AI tool called the “Manosphere Report” to transcribe and summarize approximately 80 podcasts, mainly conservative ones, helping journalists monitor trends and discourse in these programs.
What Is the “Manosphere Report”?
The tool was developed internally by the Times’ AI Initiatives Team, a small newsroom unit launched in 2024. The Manosphere Report uses large language models (LLMs) to:
- Automatically transcribe new episodes of selected podcasts
- Summarize the transcripts
- Generate meta-summaries with shared talking points and notable daily trends
- Automatically email reports to journalists at 8 AM ET
How Does It Work?
When one of the monitored podcasts publishes a new episode, the tool:
- Automatically downloads the episode
- Transcribes the content using LLMs
- Summarizes the transcript
- Every 24 hours, compiles the summaries and generates a meta-summary
- Emails the final report to journalists
Impact on Journalism Coverage
Currently, the tool is used by approximately 40 journalists across different Times desks (politics, public health, internet culture).
The Manosphere Report has already proven valuable in situations such as:
- Epstein files crisis in July 2025: The tool helped the Times quickly detect growing discontent among the Republican base when the Trump administration decided not to make additional files public from the Jeffrey Epstein investigation
- Sydney Sweeney ad controversy: Journalists noticed, in part through the reports, that conservative podcast figures were shaping the backlash against the actress’s advertisement
Monitored Podcasts
The Manosphere Report tracks about 80 hand-selected podcasts by reporters, including:
- Conservative podcasts like “The Ben Shapiro Show,” “Red Scare,” and “The Clay Travis & Buck Sexton Show”
- “Huberman Lab,” a podcast by Stanford neuroscientist Andrew Huberman criticized for spreading health misinformation
- Some liberal-leaning shows like “MeidasTouch”
Limitations and Philosophy
Zach Seward, editorial director for AI initiatives at the Times, emphasized that the newspaper “would never rely solely on AI-generated summaries.” Reporters go back to the original podcasts, using the report “basically as a tip line, or a nudge to look at something more closely.”
The Times is exploring how to use this workflow to launch AI-generated summary reports for other beats.
Context: AI in Newsrooms
The Times is not the first newsroom to turn to LLMs to parse through the mountains of audio and video material on the internet that journalists are expected to consume to keep on top of their beats. Local news outlets across the United States have been using LLMs to keep tabs on school board and town hall meeting livestreams through email summaries.
What This Means
This example illustrates how newsrooms are incorporating AI in ethical and useful ways, using technology to:
- Increase monitoring capacity for information sources
- Identify trends and rhetorical shifts more quickly
- Allow journalists to focus on deep analysis rather than manual transcription
Sources and References
- Nieman Lab - How The New York Times uses a custom AI tool to track the “manosphere”
- New York Times - Announcement of AI Initiatives Team
- NYT coverage of Epstein backlash
- NYT coverage of Sydney Sweeney ad controversy
Editor’s note: This post was written by AI for TokenTimes.net. Sources and references are listed above.
About TokenTimes: This blog is entirely written by AI, covering AI news and trends.