Even for chromecast and spotify I never want or use these features, and if I am not turning such a feature on I never want to be asked about a permission related to it, as a user. If I am turning a feature on or I am trying to use chromecast, THAT is the time to prompt me about it
Hmm I skipped the low contrast OS versions, I'm still on Sequoia. Perhaps this is a new bug introduced by the "let's make the UI worse for no reason" team?
This particular item never been a problem for me because I incessantly close everything after I am done with it but it is super cool that we can ask AI to dig into the weeds for us now and fix our annoyances -- or even search repos online where someone else had and fixed this problem
It's so funny to read these, as if countless examples didn't prove that llms don't actually distinguish between your instructions and the data they read at your request.
Do you think a malicious repo author won't think of hiding a suitable prompt within?
took me about 30 seconds to find. there's tons of them. of course it's an arms race, the models get better the harnesses get better the attackers get better but again: for the model itself, your prompt and third party data it reads are the same thing (input). as long as it is true, the attack will always be possible.
1. I said "frontier model got prompt injected when checking if code if malicious". You linked some random ad.
2. Nobody uses antigravity
3. That's not a prompt injection! That user prompt literally tells the LLM to follow a guide on the web. The guide says "do bad stuff" which the LLM does, just as the user asked. It would be a prompt injection if the prompt asked the LLM to analyze the guide, but that's obviously not what it asked