Yes.
> And with each new update you're re-evaluate?
No. Hope that helps.
I've never had a case of this where the access was malicious, and devices on my network have security controls. For me this popup is exceptionally low value.
sudo defaults write com.apple.network.local-network AllowedEthernetLocalNetworkAddresses -array "169.254.0.0/16"
sudo defaults write com.apple.network.local-network AllowedWiFiLocalNetworkAddresses -array "169.254.0.0/16"Same with MDM agents. Ours are installed automatically as part of the Apple Business Manager configuration, we manually grant them precise location in the background, and months later "HEY USER, WANNA BLOCK LOCATION FROM 'MDM SERVICE'?".
Edit: To avoid replying to the two following comments with the same reply, my gripe is not about the notification, that's a good thing; my gripe is with their implementation, which seems very in tune with Apple's way of complying with regulations. There are definitely many user-friendly ways to handle privacy compared to seemingly random notifications.
Adobe programs ask for this every time they start, and Claude Code randomly asks me this daily out of nowhere. It doesn't matter how many times I click "Don't Allow", it never gets remembered. (But of course if I ever mistakenly click "Allow" even once, that gets remembered for eternity and I never get asked again...)
I'm sure this one probably isn't, but I have no clue how I would know whether it is without reviewing the source code myself (or pointing an agent at it etc)
Do you think a malicious repo author won't think of hiding a suitable prompt within?
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.
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.
And it should be default.
OH DEAR GOD that explains so much other cursed behavior, including the "do you want to allow google chrome helper to show notifications" always reappearing. or teams. or anything that does autoupdates.
> Nothing on a typical Mac needs this permission except a printer or scanner utility, a media server, or a LAN-discovery tool, and those you answer once and they're done.
This is wrong. Browsers and media players need it because that is how multicast discovery for stuff like Chromecast works. Spotify, same thing. And Steam plus virtually all games with LAN multiplay for the same reason. Did I note already that Steam also has an autoupdater and so does Spotify and... argh.
Apple really needs to fix the very first point, preferably in a way that doesn't force app developers to ship a "helper" binary.
(Side note: it might even be the case that opening the file picker may be enough, if you got network-discoverable paths in your Finder Locations sidebar, but i'm too tired to test that one out...)
It’s weird they haven’t just made this a menu flag…
I know it's well intentioned but I have this weird setup where I am almost constantly keeping vnc on from my laptop to my desktop. And the mbpro randomly wakes up to do whatever, even when not used, which reactivates the vnc connection, which pops up a warning on my desktop, which steals my focus.
Anyone knows a way to disable that? Possibly just for connections from a specific other machine.
Edit: as for this app, I think the OS only asks once for a new app? I don't remember being asked a second time even when denying it.
I could also whine a bit about "this app wants to access files on a removable volume" since my desktop has a thunderbolt ssd plugged in for extra space...
You can see other people with tens of instances of the same app in the settings for this
A better experience would be an async consent screen, with placeholder content and calls to action until permission is granted.
For example, a network ping scanner could show placeholder host names & status until the user clicks to allow access.
The operating system could easily stub the sensitive APIs with placeholder content and a flag indicating the content is false. So even legacy apps could work.
(not saying this is ai or not, i have no idea, just speaking generally)
We're getting free from our software!
How do I know whether to say "Allow" or "Disallow" when it gives me no idea what either option actually means in concrete?
Then it started prompting users before allowing them to paste, and then the claims of privacy violations began to come in.
If instead I were able to specify what was happening in the prompt rather than iOS giving users an ambiguous statement, or even provide a detection pattern, it wouldn't have been nearly as bad.
I feel like this really harms security and privacy awareness by flooding the field with unimportant prompts/warnings to the degree that a real issue is similarly ignored because it is assumed to just be noise.
No. They’re correct, there.
The app developer should not get to provide the message text for a prompt like that.
A secondary note somewhere in less than ten words, maybe, where it explicitly warns that the statement may be false.
In App Store apps, where these messages get human review, it could more safely be allowed.
I don't really even care what OP's good intentions are, because the minute the app is sold, those good intentions are out the window. I use very few apps because I assume they are always going to potentially be adversarial to my interests.
The system warning is correct and should not be subject to the developer promising free candy to gullible users.
The fact that there are so many popups should mean apps are reaching for access to your system way beyond what they should be. It shouldn't mean Apple needs to get rid of the pop-ups.
Yes, I don't know why the OS isn't tracking the pop-up by bundle id or anything. It's not like I changed my mind the last dozen times…
before that, facebook was stealing everything on your clipboard and send back to their server.
This dialog is what it take to stop them.
and it have the pattern detect you asked.
Maybe the dialogs api should be treated like autoplaying videos with sound on the web, just disable the api if it is called outside of a click event handler, and let the consumer app break. That would be a breaking change for OSs tho. It wouldn’t prevent malicious UI patterns from triggering annoying permission prompts, but it would minimize accidental bad ux eg calling it in the main method on load
Which reminds me. Why does the CVS app need access to Bluetooth?
Then you have to require these permission controls to tell the app the user denied permission, and not just giving a valid but empty result. Some people on HN feel strongly that when you deny contacts permission to an app, the app should receive a list of dummy contacts. Without it a bad app can lock its UI and not do anything unless the permission is granted.
If you have an Extracare card, in some markets, you can self-unlock the locked display cabinets with the CVS app (because they have your identity via the ExtraCare card which adds inherent trust).
It's powered by the Invue smartlock platform. https://invue.com/smart-locks
https://abc7ny.com/post/cvs-testing-new-nyc-shoppers-open-ca...
https://youtu.be/zS9U3Gc832Y?si=BmVOHE6ahKLbw82c
Yes! This is part of the Flockification of everything.
https://indyme.com/solutions/freedom-case
The CV footage just adds that extra comfortable feeling, doesn't it?
https://indyme.com/hubfs/E6A61CF7-E252-4994-91E1-0FC0E88781F...
It's "freedom" for "honest shoppers"!