I can’t see how AI can’t be done in a privacy-respecting way [edit: note the double negative there]. The problem that worries me is performance. I have used texto-to-speech AI and it absolutely destroys my poor processors. I really hope there’s an efficient way of adding alt text, or of turning the feature off for users who don’t need it.
If it runs locally then no data ever leaves your system, so privacy would be respected. There are tons of good local-only LLMs out there right now.
As far as performance goes, current x86 CPUs are awful, but stuff coming out from ARM and likely from Intel/AMD in the future will be much better at running this stuff.
I can’t see how AI can’t be done in a privacy-respecting way [edit: note the double negative there]. The problem that worries me is performance. I have used texto-to-speech AI and it absolutely destroys my poor processors. I really hope there’s an efficient way of adding alt text, or of turning the feature off for users who don’t need it.
If it runs locally then no data ever leaves your system, so privacy would be respected. There are tons of good local-only LLMs out there right now.
As far as performance goes, current x86 CPUs are awful, but stuff coming out from ARM and likely from Intel/AMD in the future will be much better at running this stuff.