Website "In the Weights" shows whether AI models know who you are
Two ex-OpenAI engineers just built a tool that scores how deeply you're embedded in AI training data. Mozart: 996. You: probably lower.
Why it matters
A new public tool reveals the opacity of AI training data and which individuals are most represented in foundation models—raising questions about data provenance, consent, and model behavior bias toward well-known figures.
The key facts
11 to knowTool name: 'In the Weights'
Built by two former OpenAI employees
Reveals training data recall of individuals across AI models
Strength score scale: up to 996
Top-ranked individuals: Mozart, Shakespeare, Taylor Swift
Implicit privacy/transparency angle on training data composition
Tool built by former OpenAI employees
Strength scoring system up to 996
Top figures: Mozart, Shakespeare, Taylor Swift
Reveals memorization/recall from training data
Public-facing web application
Go to the source
The Decoderthe-decoder.com
Publisher excerpt: Two former OpenAI employees have built a website called "In the Weights" that reveals which people AI models can recall purely from their training data. A strength score of up to 996 shows how deeply a person is embedded, with Mozart, Shakespeare, and Taylor Swift topping the list.