# Inside ‘Project Lily’: The Humans Reading Your ChatGPT Chats

> A 404 Media investigation lifts the lid on Project Lily, OpenAI's internal effort in which hundreds of contract workers read real ChatGPT conversations and rate the responses, reportedly on a scale of one to seven. The stated goal is to improve the models, including tamping down excessive flattery and overly human-like behavior, but the practice means human strangers are looking at prompts that can contain deeply personal and sensitive information. The consent story is the uncomfortable part. According to the reporting, the setting that feeds conversations into this review, labeled something close to improving the model for everyone, is switched on by default, so users who assumed their chats were private have to actively opt out to keep humans from reading them. It is a concrete reminder that behind the seamless chatbot experience sits a familiar apparatus of human labeling, one that most people never see and did not knowingly agree to. As chatbots become confidants for medical, legal, and emotional questions, the gap between what users assume and what actually happens with their words is only going to matter more.

_Section: [Daily AI Updates](https://www.wortins.com/daily-ai) · Source: 404 Media · Published Tuesday, September 15, 2026_

## Wortins' read

A 404 Media investigation lifts the lid on Project Lily, OpenAI's internal effort in which hundreds of contract workers read real ChatGPT conversations and rate the responses, reportedly on a scale of one to seven. The stated goal is to improve the models, including tamping down excessive flattery and overly human-like behavior, but the practice means human strangers are looking at prompts that can contain deeply personal and sensitive information. The consent story is the uncomfortable part. According to the reporting, the setting that feeds conversations into this review, labeled something close to improving the model for everyone, is switched on by default, so users who assumed their chats were private have to actively opt out to keep humans from reading them. It is a concrete reminder that behind the seamless chatbot experience sits a familiar apparatus of human labeling, one that most people never see and did not knowingly agree to. As chatbots become confidants for medical, legal, and emotional questions, the gap between what users assume and what actually happens with their words is only going to matter more.

## Source

[Read the full story at 404 Media](https://www.404media.co/inside-project-lily-the-humans-reading-your-chatgpt-chats/)

## Related coverage

- [OpenAI researcher: top models are becoming so situationally aware humans "are losing the ability to evaluate them" while humans rely more on AI to lead research (Daniel Kokotajlo/@dkokotajlo)](https://www.wortins.com/story/openai-researcher-top-models-are-becoming-so-situationally-a-a332594c) — [Techmeme](https://www.techmeme.com/260914/p45#a260914p45)
- [Trump facing AI backlash in Congress as push for guardrails intensifies](https://www.wortins.com/story/trump-facing-ai-backlash-in-congress-as-push-for-guardrails--039bf497) — [The Guardian](https://www.theguardian.com/technology/2026/sep/15/trump-ai-guardrails-democrats-republicans)
- [The real AI economy is being built by ordinary people](https://www.wortins.com/story/the-real-ai-economy-is-being-built-by-ordinary-people-41e703e0) — [Rest of World](https://restofworld.org/2026/global-workers-using-ai-at-work/?utm_source=rss&utm_medium=rss&utm_campaign=feeds)
- [US plans to raise AI-directed cyberattacks with China during Xi summit](https://www.wortins.com/story/us-plans-to-raise-ai-directed-cyberattacks-with-china-during-55d719a2) — [The Next Web](https://thenextweb.com/news/us-china-ai-talks-cyberattacks-trump-xi-summit-september)
- [Microsoft’s new AI rulebook rejects the research its $5bn partner is doing](https://www.wortins.com/story/microsoft-s-new-ai-rulebook-rejects-the-research-its-5bn-par-3c9ebd6d) — [The Next Web](https://thenextweb.com/news/microsoft-ai-code-of-conduct-model-welfare-anthropic)
- [OpenAI urges UK lawmakers to rein in technology amid growing safety fears](https://www.wortins.com/story/openai-urges-uk-lawmakers-to-rein-in-technology-amid-growing-8448a970) — [The Guardian](https://www.theguardian.com/technology/2026/sep/14/ai-regulation-anthropic-uk-human-rights-committee-mps-lords)

---

_Curated and written by [Wortins](https://www.wortins.com) — The daily AI briefing. Every story links to its original source; the "Wortins read" on each is our own original analysis. [About Wortins & our editorial approach](https://www.wortins.com/about)._