# Meta's Fake Teen Factory

Page: https://stenobird.com/podcast/generative-ai-101-6932184/meta-s-fake-teen-factory
Text version: https://stenobird.com/podcast/generative-ai-101-6932184/meta-s-fake-teen-factory.md
Podcast: [Generative AI 101](https://stenobird.com/podcast/generative-ai-101-6932184)
Published: 2026-07-13T10:13:00+00:00
Episode link: https://generativeai101.podbean.com/e/metas-fake-teen-factory/
Audio file: https://mcdn.podbean.com/mf/web/h42j89sqeaae5xse/Meta.mp3
Processing state: not_requested
JSON: https://stenobird.com/v1/public/podcasts/generative-ai-101-6932184/episodes/meta-s-fake-teen-factory
Duration seconds: 446

## Resource

Meta paid contractors to pose as children, flood rival chatbots with prompts about suicide, eating disorders, and abuse, then log every response in spreadsheets. The company calls it "industry-standard safety benchmarking," but the operation had no consent, no disclosure, and no shared findings: the four things that make red teaming legitimate. Host Emily Laird walks through the Wired investigation, the 45,000-prompt testing rounds, and the court testimony showing what Meta knew about its own failure rates while it was busy documenting everyone else's. This is the difference between a shield and a sword, and the paperwork says sword.──────────── 🎯 JOIN THE AI WEEKLY MEETUPS https://www.uwstout.edu/ai-weekly-meetup 📩 EMAIL REMINDERS FOR THE MEETUPS https://app.e2ma.net/app2/audience/signup/2101263/1779703/ 💬 CONNECT WITH EMILY LAIRD ON LINKEDIN http://www.linkedin.com/in/meet-emily-laird

## Actions

- request_transcript: `POST https://stenobird.com/v1/public/podcasts/generative-ai-101-6932184/episodes/meta-s-fake-teen-factory/transcription-requests` — Idempotently request low-priority transcript generation for this episode.
- read_markdown: `GET https://stenobird.com/podcast/generative-ai-101-6932184/meta-s-fake-teen-factory.md` — Read the agent-friendly Markdown representation of this episode resource.

A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed.

## Transcript

Full transcripts are not published on public pages unless there is a clear rights basis.