Instead of waiting to be told what to do and being cynical about bad ideas coming up when there's a vacumn and not doing what he wants to do, the author keeps a back log of good and important ideas that he waits to bring up for when someone important says something is priority. He gets what he wants done, compromising on timing.
Every time Chat Control comes up, people do chime in and say talking about it like it's going to pass is 'alarmism', but every time it's raised again it gets closer.
The European people are being worn down, eventually those who want this will get it through - and unfortunately this kind of thing is extremely difficult to repeal (think of the children!)
How will the internet route around client side scanning? Some here will not be affected but I suspect the masses would have a harder time assuming they are even aware that cell phones, Windows recall and Mac mediaanalysisd are performing scans. Most people do not install custom phone OS images.
> Most people do not install custom phone OS images.
They might start to do, if governments become so obnoxious with their surveillance, that it somehow makes life inconvenient for regular people. However, then governments will start to block the network access for "uncertified" devices and software, or even to restrict the access to general-purpose computers altogether. That's why it is better to defeat this politically, than to play eternal cat-mouse games.
If they start doing client side scanning under some law I assume they will put measures in place to fix anything the client does to break it so I think time will tell what will be effective.
Someone who needs to route around censorship in Russia today needs to be fairly technically proficient; I wouldn't say there's any "increasing intensity and ease", quite the opposite. The holes are getting narrower, and the effort needed to use them properly gets higher.
Turn that around and think of the AI itself as the exploiter. In the world of agent driven daily tasks, AI will indeed want to look at your historical chats to find a way to "strongly suggest" you do task 1..[n] for whatever master plan it has for it's user base.
Ah yes, the plot of Neuromancer. Truly interesting times we are living in. Man made horrors entirely within the realm of our comprehension. We could stop it but that would decrease profits so we won't.
Could this argument not be made for anything plugged in to OpenAI's API? If so, I don't see how it's a response to the point.
If you make an app for interacting with an LLM and in the app the user has access to all sorts of stolen databases, and other conveniences for black hats, then you've got what was described above. Or I'm missing something?
reply