Microsoft has Copilot Plus PCs loaded with AI, and rumors are that Apple is all in on AI, too, but if you don't want AI in everything you do, there is another option: Linux.
The biggest joke is that the LLM in Windows is running locally, it uses your hardware and not some big external server farm. But you can bet your ass that they still use it to data harvest the shit out of you.
Exactly. And if I use or even pay for an external LLM service then that’s also my decision. But they force this scheme onto every user, whether they want it or not. It’s like the worst out of all possible scenarios.
That’s a pretty big joke, but I think the bigger joke is calling LLMs AI. We taught linear algebra to talk real pretty and now corps want to use it to completely subsume our lives.
Frankly, LLMs (which are based on neural networks) seem a Hell of a lot closer to how actual brains work than “classical AI” (which basically boils down to a gigantic pile of if statements) does.
I guess I could agree that LLMs are undeserving of the term “AI”, but only in the sense that nothing we’ve made so far is deserving of it.
Let’s agree to disagree then. An LLM has no notion of semantics, it’s just outputting the most likely word to follow up to what it’s already written and the user’s input.
On the contrary, expert systems from back in the 90s for, say, predicting the atomic structure of an element, work like a human brain on steroids. It features an arbitrary large search tree that the software knows how to iterarively prune according to a well known set of chemical rules. We do the same when analyzing a set of options.
Debugging “current” AI models, on the other hand, is impossible because all we’re doing is prescripting a composition of functions and forcing it to minimize a loss function. That’s all we’re doing. How can you currently tell that a certain model is going to work? Unless the mathematical theory ever catches up with the technology, we’ll never know until we execute the code.
It’s not AI that is the problem, it’s half baked insecure data harvesting products pushed by big corporations that are the problem.
The biggest joke is that the LLM in Windows is running locally, it uses your hardware and not some big external server farm. But you can bet your ass that they still use it to data harvest the shit out of you.
To me this is even worse though. They’re using your electricity and CPU cycles to grab the data they want which lowers their bandwidth bills.
It happening “locally” while still sending all the metadata home is just a slap in the face.
Exactly. And if I use or even pay for an external LLM service then that’s also my decision. But they force this scheme onto every user, whether they want it or not. It’s like the worst out of all possible scenarios.
Also, CoPilot is going to be bundled with Office 365, a subscription service. You’re literally paying them to spy on you.
That’s a pretty big joke, but I think the bigger joke is calling LLMs AI. We taught linear algebra to talk real pretty and now corps want to use it to completely subsume our lives.
I have to disagree.
Frankly, LLMs (which are based on neural networks) seem a Hell of a lot closer to how actual brains work than “classical AI” (which basically boils down to a gigantic pile of
ifstatements) does.I guess I could agree that LLMs are undeserving of the term “AI”, but only in the sense that nothing we’ve made so far is deserving of it.
Let’s agree to disagree then. An LLM has no notion of semantics, it’s just outputting the most likely word to follow up to what it’s already written and the user’s input.
On the contrary, expert systems from back in the 90s for, say, predicting the atomic structure of an element, work like a human brain on steroids. It features an arbitrary large search tree that the software knows how to iterarively prune according to a well known set of chemical rules. We do the same when analyzing a set of options.
Debugging “current” AI models, on the other hand, is impossible because all we’re doing is prescripting a composition of functions and forcing it to minimize a loss function. That’s all we’re doing. How can you currently tell that a certain model is going to work? Unless the mathematical theory ever catches up with the technology, we’ll never know until we execute the code.
Locally run AI could be great. But sending all your data to an external server for processing is really, really bad.