what application are you trying to achieve? Even the example of spying in this article isnt applicable
If I hired a private detective to spy on you, that detective could hide a bug in your home or car, tap your phone, and listen to what you said. At the end, I would get a report of all the conversations you had and the contents of those conversations.
you wont get a summarized report, you get transcripts, you get full emails, you get audio files, if you’ve ever followed investigations. It’s not an impossibly large stack of communication from someone saying or typing a million words a day. It’s very easily filterable and searchable.
This allows something like “computer show me every citizen that has posted negative comments towards the current mayor, order it by the amount of comments theyve posted, cite where they work, who their parents are, what people does there phone gps spend the most time around, and include any cctv and amazon ring footage that detects their face using facial recognition, create a separate folder of footage from any bar and club they’ve been to, search every nsfw website/subreddit/onlyfans for footage of them.”
And now you have a neatly organized collection of all the people that oppose the mayor with potential blackmail and who to send it to one by one until your demands of only praising the great cop mayor is met (as an example)
yeah but again there are human limitations. Each is filtered independently but with ai they can be correlated. Its much like how chatbots are useful as the next stage of search. Search itself went from library and texts to online but when results started giving a brief sumary it greatly enhanced the speed someone can go through the links and now with chatbots its essentially doing the link result search for you. As a human you have to go through it to verify but you can get to results much faster.
Theyre saying AI can summarize. Thats it.
that is pretty huge though. right now the limitation of large datasets is the inability for someone to sift through it.
what application are you trying to achieve? Even the example of spying in this article isnt applicable
you wont get a summarized report, you get transcripts, you get full emails, you get audio files, if you’ve ever followed investigations. It’s not an impossibly large stack of communication from someone saying or typing a million words a day. It’s very easily filterable and searchable.
This allows something like “computer show me every citizen that has posted negative comments towards the current mayor, order it by the amount of comments theyve posted, cite where they work, who their parents are, what people does there phone gps spend the most time around, and include any cctv and amazon ring footage that detects their face using facial recognition, create a separate folder of footage from any bar and club they’ve been to, search every nsfw website/subreddit/onlyfans for footage of them.”
And now you have a neatly organized collection of all the people that oppose the mayor with potential blackmail and who to send it to one by one until your demands of only praising the great cop mayor is met (as an example)
yeah but again there are human limitations. Each is filtered independently but with ai they can be correlated. Its much like how chatbots are useful as the next stage of search. Search itself went from library and texts to online but when results started giving a brief sumary it greatly enhanced the speed someone can go through the links and now with chatbots its essentially doing the link result search for you. As a human you have to go through it to verify but you can get to results much faster.