social.tchncs.de is one of the many independent Mastodon servers you can use to participate in the fediverse.
A friendly server from Germany – which tends to attract techy people, but welcomes everybody. This is one of the oldest Mastodon instances.

Administered by:

Server stats:

3.7K
active users

#SafetyCore

0 posts0 participants0 posts today

Un monde où l'on ne peut pas faire confiance à ses outils de communication, où l'on doit s'échanger entre initié des techniques pour juste désactiver une truc espion *de plus* dans son téléphone, n'est pas un futur souhaitable.

Android installe silencieusement un logiciel pour vous "protéger" à l'insu de votre plein gré
⤵️
frandroid.com/android/2505247_

#Android
#SafetyCore
#TousSuspects
#Surveillance

Replied in thread

@Tutanota Непогана стаття про альтернативні мессенджери, але зовсім не відкрили тему #Android #Google #Messages це гуглосміття без попередження та запиту встановлює з 2025 року пару програмуль на ваші мобільні пристрої – слідкують заради вошого добробуту! Раджу перевірити пристрої та видалити:
1) Android System #SafetyCore
2) Android System #KeyVerifier

"Google installiert einen Nacktscanner auf deinem Handy – das steckt dahinter" - von @schurt3r.bsky.social @watson.ch

Lohnt sich, sich damit zu beschäftigen

Bei HyperOS 2.0 (Xiaomi) sind die Berechtigungen von standardmässig deaktiviert

watson.ch/digital/google/91464

watson · Google installiert Nacktbild-Scanner auf Android-Handys – das wissen wirBy Daniel Schurter

Bereits seit Oktober 2024 liefert #Google für #Android 9+ eine Komponente namens "Android System #SafetyCore" aus, die wohl mittels KI u.a. Spam und Nacktbilder in Messages erkennen kann, deinstalliert werden kann, aber nach Berichten teils auch wieder installiert wird.

Datenflüsse nach außen sind bislang weder für die Komponente noch für das verwendete (und lokale) KI-Modell bekannt.

#Graphene-Nutzende bekommen den Kram nicht.

ZDNet berichtet hier: zdnet.com/article/a-new-androi

Update: fixed typo

ZDNET · A new Android feature is scanning your photos for 'sensitive content' - how to stop itBy Steven Vaughan-Nichols

Have you noticed there's a lot more drama going on here lately?

@finn and I did at least, so we wrote a blog post talking about what happened with VLC and Google, how that drama formed and what you can do to stop spreading misinformation.

steffo.blog/outrage-warps-real

This is my first-ever blog post that I wrote with someone together, so I hope the way I show what's written by whom is understandable. :neofox_laugh_sweat:

Anyway, thank you, Finn, for helping me write such an important blog post. I think I couldn't have done it without you! :floofHeart:

Steffo Blog · Outrage Warps Reality
More from Steffo :steffo:

It's really getting on my nerves that there's drama about something that I'd normally only see in app reviews on the Play Store. Like... In the Play Store I sometimes look at the (German) reviews of some pre-installed apps and laugh my ass off because holy shit it's so funny how stupid people are. ("I DON'T WANT THIS THEY WANT TO SPY ON ME AAA")

Now that there's actually many posts on fedi talking about how an pre-installed app spies on the users and scans images.... no? Haven't you guys read what it does?
Oh, I forgot. We're on fedi, where stuff like that starts and then just gets bigger - without thinking about it.

Sure, boosts can be well-intentioned, but without checking the information first, a boost can lead to a lot of FUD. (By the way, I also consider myself guilty here.)

It started getting on my nerves with the whole "VLC uses AI!! They want to kill all subtitles!! USE SOMETHING DIFFERENT 😱" thing, but now it's happening again??

So, what is that freshly new app "Android System SafetyCore" actually do?

Apparently it's just a collection of *local AI models* that can classify stuff *locally*, *without sending it to Google*. These can then be used by other apps, so they don't have to send it to other services / ship their own ai models.
Do you want an example for this? Google Messages uses this new app to classify messages as spam, malware, nudity, etc.

I can recommend reading this thread if you want to know more about this new app: grapheneos.social/@GrapheneOS/

But please... PLEASE check before going code red and screaming on the internet that something is bad, because, again, that's something I always made fun of. Because this shows me that you can't read. (Or check actual sources.)

Thanks.

GrapheneOS MastodonGrapheneOS (@GrapheneOS@grapheneos.social)The app doesn't provide client-side scanning used to report things to Google or anyone else. It provides on-device machine learning models usable by applications to classify content as being spam, scams, malware, etc. This allows apps to check content locally without sharing it with a service and mark it with warnings for users.