What is Nostr?
Lisa Kalayji /
npub1tq0…3w5p
2024-11-26 18:11:42

Lisa Kalayji on Nostr: I work for a legal aid organization. What do you think happens to attorney/client ...

I work for a legal aid organization. What do you think happens to attorney/client privileged info when MS swipes it straight out of documents and makes it possible for anyone and everyone to find that sensitive info just by typing the right prompts into the LLM?

"The company seems to have quietly begun “scraping” AI training data from information people are typing in to its desktop publishing program Word and its spreadsheet tool Excel...

This has gigantic implications for anyone who may use these well-known tools... The issue is that when a piece of content is used to train one of the current crop of AIs—like Microsoft’s Copilot—an imprint, of sorts, is left in the algorithm. When a different user asks for information at some point in the future, it’s possible that your content, however sensitive it may be, or whatever proprietary information it contains, may “surface” in some form in the generative AI’s response."

https://www.inc.com/kit-eaton/after-office-software-outage-microsoft-faces-allegations-of-data-scraping-in-word-and-excel/91024539

#Privacy #Tech #Data #Microsoft #AI #OnlineSafety #CivilRights
Author Public Key
npub1tq0f72xp46j5a9q6q5m9627un96l2wvzut0ncvvkcxstc4r6h8wqj73w5p