Microsoft’s Recall feature uses AI to capture and store user data. While it can be useful, it also poses a significant privacy risk. Here's what AdGuard is doing about it.
It’s a centralized search that can dig through your activity cross-platform and parses it through a centralized AI. Whether the data is stored in a log or as screenshots is a difference, but not as big of a difference as people make it out to be. It just feels intuitively weirder because one is humanly readable and the other one isn’t.
To be fair, that’s my takeaway from a lot of AI backlash. A whole bunch of it is people finally getting an intuitive grasp on activities that big data has been doing for years or decades and it finally clicking into shock because they can anthropomorphise the inputs and outputs better.
No wonder the techbros have lost their intuititon for what will trigger backlash. In many cases they’ve been doing far worse than those things with zero awareness or pushback.
Access controls is the big difference. Apps with sensitive data can choose to hide stuff to a system wid search API. It can do so on an individual level, even. And even if it previously was accessible it can be drumrollrecalled. Exposure happens when a search is made.
Microsoft Recall is all or nothing. Once it has been displayed Recall has it and you can’t selectively erase stuff. Exposure is immediate. It’s just purge the whole database, or leave it all in there. Apps can’t retroactively flag stuff.
… But leaving AI summaries on by default was very stupid by Apple
I’d argue that this is way more nuance than the public in general puts into the issue. In fact, the goalposts have moved quite a bit. “The big difference” used to be the local encription of the data, but it became not it once Recall implemented that. Or the opt-in, which went the same way.
That’s not to say I don’t think it’s a better idea to have per-app support (which is incidentally how Microsoft implemented the feature in Windows 8 the first time), but I will say that’s not why people are mad at one and not the other.
I don’t actually know if you can selectively erase specific screenshots from the database because I, again, can’t find any traces of Recall on my supported PCs for the life of me. Coverage had made it seem that they could, since presumably the much criticised side effect of having a local, freely accessible database with just a bunch of pictures is that you could… you know, access those. Did they obscure it further in the reimplementation?
And also, I think people believe I’m being argumentative, but I’m not. Can somebody point me at the Recall opt-in and/or some explanation why my Copilot + device running 24H2 would not seem to have it available anywhere? I’m confused about the rollout here. I don’t want it on, but I’d like to try it and see what the practical implementation is for myself (and be double sure I have it turned off once I’m done with that).
After the heavy criticism they changed it from default on to (opt out) to default off (opt in).
In theory you could modify it’s database, but they did mention applying stricter security (but what good does that do when the frontdoor access remains via the prompts)
Yeah, and that didn’t change the perception people have of it. That’s the point I’m making.
And I’ll get back to you on how easy it is to specifically remove specific data entries (and whether or not the prompts are handled locally or remotely) if and when I can get a hold of this mythical unicorn, because by how much people ignores my questions I’m assuming nobody here has actually tried it?
It’s a centralized search that can dig through your activity cross-platform and parses it through a centralized AI. Whether the data is stored in a log or as screenshots is a difference, but not as big of a difference as people make it out to be. It just feels intuitively weirder because one is humanly readable and the other one isn’t.
To be fair, that’s my takeaway from a lot of AI backlash. A whole bunch of it is people finally getting an intuitive grasp on activities that big data has been doing for years or decades and it finally clicking into shock because they can anthropomorphise the inputs and outputs better.
No wonder the techbros have lost their intuititon for what will trigger backlash. In many cases they’ve been doing far worse than those things with zero awareness or pushback.
Don’t worry, Microsoft is bringing semantic search to Windows too. That way you can have the worst of both worlds.
Access controls is the big difference. Apps with sensitive data can choose to hide stuff to a system wid search API. It can do so on an individual level, even. And even if it previously was accessible it can be drumroll recalled. Exposure happens when a search is made.
Microsoft Recall is all or nothing. Once it has been displayed Recall has it and you can’t selectively erase stuff. Exposure is immediate. It’s just purge the whole database, or leave it all in there. Apps can’t retroactively flag stuff.
… But leaving AI summaries on by default was very stupid by Apple
I’d argue that this is way more nuance than the public in general puts into the issue. In fact, the goalposts have moved quite a bit. “The big difference” used to be the local encription of the data, but it became not it once Recall implemented that. Or the opt-in, which went the same way.
That’s not to say I don’t think it’s a better idea to have per-app support (which is incidentally how Microsoft implemented the feature in Windows 8 the first time), but I will say that’s not why people are mad at one and not the other.
I don’t actually know if you can selectively erase specific screenshots from the database because I, again, can’t find any traces of Recall on my supported PCs for the life of me. Coverage had made it seem that they could, since presumably the much criticised side effect of having a local, freely accessible database with just a bunch of pictures is that you could… you know, access those. Did they obscure it further in the reimplementation?
And also, I think people believe I’m being argumentative, but I’m not. Can somebody point me at the Recall opt-in and/or some explanation why my Copilot + device running 24H2 would not seem to have it available anywhere? I’m confused about the rollout here. I don’t want it on, but I’d like to try it and see what the practical implementation is for myself (and be double sure I have it turned off once I’m done with that).
After the heavy criticism they changed it from default on to (opt out) to default off (opt in).
In theory you could modify it’s database, but they did mention applying stricter security (but what good does that do when the frontdoor access remains via the prompts)
Yeah, and that didn’t change the perception people have of it. That’s the point I’m making.
And I’ll get back to you on how easy it is to specifically remove specific data entries (and whether or not the prompts are handled locally or remotely) if and when I can get a hold of this mythical unicorn, because by how much people ignores my questions I’m assuming nobody here has actually tried it?