All recent OSes even visually make it obvious that the microphone is used, after asking the user if they should allow it at all. I don’t think they can actually to more far than this. Maybe regularly ask the user if they really want the microphone to be used all the time.
It isn’t better, it’s even worse IMO. But it’s just the truth. It is verifiable enough that apps cannot just spy via your microphone in any reasonably modern OS, not even the old versions. What has never been verified though is that there are non-zero-day-exploit-ways to spy through your microphone.
Bro, they don’t. The OS manages microphone access. If you allow it, yes, they may spy. But even then, the OS visible tells you via an icon that the microphone is used. It is the plain written truth, that we can not argue around. There has been a lot security research around this, stuff has been audited, Android is even open source, and no one has ever found even a hint of this being possible.
Echos and Google Homes are an entirely different story because they operate their microphones by design, their entire system works on always listening, and they don’t hide that.
PLEASE PLEASE split WhatsApp from Meta. I can’t get my friends off it and lately the integration of this Llama button + automatic Llama search when I’m just looking for a chat really reminds me that I should not and should’ve never used this app.
It’s not like they’re using an LLM like ChatGPT. Science and research has used AI for years and it has a very big potential.
Most people nowadays mistake AI as LLM, and aren’t aware of the scientific fields where AI has been and is used, and has been used for way longer than LLMs are publicly known.