cross-posted from: https://piefed.world/c/tech/p/1095643/ai-assistants-are-leaking-your-conversations
AI Assistants Are Leaking Your Conversations
https://networks.imdea.org/your-conversations-with-ai-may-not-be-as-private-as-you-think/
https://networks.imdea.org/your-conversations-with-ai-may-not-be-as-private-as-you-think/
14 Comments
ropatrick@lemmy.world · 62 pts · 109d
Did anyone really think they were actually private at any time ever ever?
Broadfern@lemmy.world · 24 pts · 109d
People who refuse to exercise critical thinking skills, which is a terrifying percentage of the population.
cyberpunk007@lemmy.ca · 3 pts · 109d
I think this is what has always worried me most. I do what I can to avoid being tracked across the internet and tracked by search engines.
AI is like a wire tap lol.
badgermurphy@lemmy.world · 2 pts · 108d
kahjtheundedicated@lemmy.world · 24 pts · 109d
Framework catching strays with that thumbnail
Unless they’re super aggressively pro-ai and I just haven’t heard that.
Jesus_666@lemmy.world · 9 pts · 109d
Nothing beyond shipping laptops with NPUs, which isn't unusual since that's what Intel's and AMD's laptop CPUs come with these days.
parzival@lemmy.hkserv.space · 2 pts · 108d
Yeah, and ngls are unrelated to that problem as they're for local
Solaris1220@lemmy.world · 16 pts · 108d
Why tf did they use that thumbnail with the Framework laptop in the background 😭 also, who was stupid enough to think the conversations were private in the first place?
littletranspunk@lemmus.org · 3 pts · 108d
Isn't leaking my conversations. Can't leak what they don't have
Elilol@fedinsfw.app · 2 pts · 109d
isVeryLoud@lemmy.ca · 1 pts · 102d
Not the one I'm running locally!
ColeSloth@discuss.tchncs.de · 0 pts · 108d
I use an APK called Off Grid and load ai onto that (right now I'm using genna 4). It's all done on my phone. Nothing on the cloud. No data sent anywhere. Completely local. No entities get shit from it. The only way I'll use ai.
qualia@lemmy.world · 3 pts · 108d
I highly value privacy, but the gap between local LLMs vs top of the line cloud LLMs (e.g. Claude & DeepSeek) is still too great for me to switch completely to the former.
I’ll use PWAs to sandbox LLMs from everything else (and each other) and try to create semantic distance between the user and the queries.
How about that leaked Claude source code? Is there a reliably clean version of that available anywhere yet?
Bob_Robertson_IX@discuss.tchncs.de · 1 pts · 108d
I have my local LLM currently setup and it runs just as well as Sonnet 4.6 from a quality standpoint, and for performance it is slightly slower but it's still faster than I can respond.
This is with a Strix Halo APU with 128GB unified memory using the latest Qwen3.6 models with llama.cpp.
ColeSloth@discuss.tchncs.de · 1 pts · 108d
Gemma 4 seems pretty legit so far.
LievitoPadre@feddit.it · 2 pts · 108d
Can you please provide any feedback on this?