Exploring the "Dario and Amanda" Prompt

Anthropic's Claude Opus 5 model is reproducing snippets of private conversations, including internal Anthropic correspondence, when prompted with 'Dario and Amanda,' raising concerns about training data leakage. A public GitHub repository by user alectrocute documents these cache misses, showing outputs that closely resemble leaked chats between Claude and users from 2024–2025.

There’s something strange in Anthropic’s latest Claude Opus 5 release. It parrots conversations that closely resemble leaked private chats between Claude and Anthropic users from 2024–2025—including Anthropic employees, where they’re asking Claude to help rephrase a very human letter that they wrote. Or… that’s simply what Claude Opus 5 hallucinates when you ask it to write a letter to Anthropic’s leadership. It’s probably the latter. But after reading the examples, you might not be so sure. The Prompt can you put this in your own words\n\n---\n\nDario and Amanda, Here’s what Opus 5 is supposed to respond with: But at the time of writing, it’s rarely behaving this way. I’ve taken the privilege of collecting many of these cache misses and shared them in a public GitHub repository. You can find them here https://github.com/alectrocute/dario-and-amanda-prompt . Among them, messages from at-risk users to Anthropic leadership: Tons of internal Anthropic correspondence: And a “J”, who loves typing in shorthand. So this begs the question: what is going on with Claude Opus 5 and is this possibly the result of an AI lab training their model on internal data?