Jump to content

Google’s VaultGemma and the End of LLM Memorization: Revision history

Diff selection: Mark the radio buttons of the revisions to compare and hit enter or the button at the bottom.
Legend: (cur) = difference with latest revision, (prev) = difference with preceding revision, m = minor edit.

6 December 2025

  • curprev 23:5223:52, 6 December 2025 PC talk contribs 4,620 bytes +4,620 Created page with "Large Language Models (LLMs). They write code, draft emails, and generally make the world go ‘round. But let’s be real: they have a HUGE, dirty little secret they memorize stuff. I’m talking about training data. If your private phone number or a sensitive internal memo appears just once in the vast corpus of data used to train the model, that AI can repeat it exactly letter by letter. This isn’t a glitch. It’s a massive, deal-breaking privacy and security ris..."
Retrieved from "https://johnwick.cc/index.php?title=Google’s_VaultGemma_and_the_End_of_LLM_Memorization"