4 ms·
For now the system has to be trained per person. Is the difference in brain activity across individuals significant enough to stop building a universal model?
by devsda 3y ago
For now the system has to be trained per person. Is the difference in brain activity across individuals significant enough to stop building a universal model?
It is not clear whether this is language dependent or not. If it is, I hope being bilingual or multilingual is a viable defense against this technology.
- sjducb 3y agoObviously this is pure speculation, but I suspect it'll be a lot like cloning someone's voice. The first systems required the participant to read thousands of lines of text. Now you can do it over a phone call without the participant noticing that you're cloning their voice.
- sycamoretrees 3y agoWhy would bilingualism be a defense? Try throwing ChatGPT or whatever a line which combines several languages. e.g., If you ask it “¿Quel est the resultado of 四*6.1, in inglese?”, it will probably reply “4 times 6.1 equals 24.4.” I suppose AI doesn’t think in words, but expects bits of information to be related to X meaning, and from there it can predict a reply and translate that information into words we understand. I haven’t tried the above line, but if it’s not possible already there’s nothing really stopping us from making it work soon enough. Edit: sorry, maybe I’m misunderstanding. Are you talking more about multilingual people thinking more abstractly, which would make it harder to tie information to any specific words?