Hacker Newsnew | past | comments | ask | show | jobs | submit | johncassil's commentslogin

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


It's a user experience thing. They don't want to show the cast icon to the user if there is no device present to cast to.

(I think both POVs are valid, but the actual solution should be for a proper OS-level interface to provide mDNS discovery...)


Regardless, apple should definitely not send popups unless an app is being engaged with.


https://forums.macrumors.com/threads/repeated-allow-google-c...

You can see other people with tens of instances of the same app in the settings for this


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?


I assume that is a quick fork and publish, but seems scary lol


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


I maintain that Steve Jobs would have taken the engineers to task over this poor UX.


Good question -- I have been pointing claude at repos for security reviews before I install stuff now for exactly this reason.


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?


> countless examples

Show me just one where a frontier model got prompt injected when checking if code if malicious.

FWIW I can do this, but I bet you can't.


Here's prompt injection via a harness for example: https://www.promptarmor.com/resources/google-antigravity-exf...

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


You are under an illusion that prompt is treated as a different kind of input by the model, than data it reads.


Still waiting. You had "countless examples", but I only asked for one. Have you never successfully prompt injected an LLM?


> as if countless examples didn't prove that llms don't actually distinguish between your instructions and the data they read at your request.

That's what I wrote


oof, it was originally one, forgot about the refactor lol


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: