Meredith Whitaker

You have, you know, the 11% of Anthropik's clawed co-pilot experiment, or clawed chrome experiment, 11% of the time, even with guardrails, it was not able to resist prompt injection.

Muse is logging people's contact lists, logging the people they talk to. So you install Muse, and I am in your contact list. Muse now has my information. They are logging when we talk to each other. They are creating dossiers about me and your communications with me, Cambridge Analytica style.

We're talking about Meta's Muse, we had a person show up to someone else's house because Muse had given that person the address to that house. Because that person said, hey, Muse, can you manage my Facebook marketplace for me? And Muse took that to mean we're giving out your address and then impersonating you when this guy stands outside your door confused, wondering why you didn't come out to get the product he was going to buy. You have Muse now publishing private information online, company r

I started working at Google in 2006 and so I watched a business model that was reliant on grabbing as much data as possible about as many people as possible using that to create models that were then sold to advertisers to enhance their business. I watched that metastasize from the inside and I became concerned about it.

We have researchers showing a prompt injection attack on Google Calendar that led an agent to turn on the boiler in someone's smart home.