I’m not saying Meta isn’t spying on users. I’m sure they are. But this doesn’t make sense as a symptom of spying.
If you want to spy on users you just record all their info. You absolutely don’t give the LLM they talk to access to that same information - that’s nearly useless, and is one of the surest ways to make sure the users find out. The LLM doesn’t need to know about the spying.
I really hate that this article is pretending that Muse is somehow “spying” on users.
This isn’t spying. This is just bad engineering.
An LLM should only be given access to relevant information for a particular query. If you give an LLM access to irrelevant information, it gets confused. Anyone who’s used LLMs much knows this. It’s kind of baffling to me that the project managers over at Meta don’t understand how dumb their design is.
Most DEs have their own specific way to do this, so the exact solution will vary.
That said, if you’re looking for a general solution, there’s good odds your computer has a BIOS setting that can be used to set the initial numlock state, and that will work no matter what DE you use. If that’s an option for you, that would be my first choice.
I feel like every story I hear out of the UK lately is some weird combination of stupid and vaguely dystopian.
Are things ok over there?
It’s worth calling out that there’s no real justification for that 2% figure. It’s a number that was pulled out of the air by the New Zealand version of the Fed in the late 80s, and it seems to work well, so everyone copied it.
There’s pretty broad consensus among economists that deflation is bad, and that inflation above about 3-4% is bad. Beyond that, we’re literally just guessing.
If I ask an LLM to review my code it will find a long list of issues. Some of them will be real, and some of them will look kind of like issues, but won’t really be when you dig into them. Sometimes the issues will be kind of real, in the sense that there’s a real best practice not being quite followed, but they don’t actually rise to the level of a bug or vulnerability, and they’re not worth fixing.
If I fix that first list of issues, then ask an LLM to review my code again, it will, almost without fail, generate an equally long list of issues the second time. There might be some new real issues it missed in the first pass, but it will be much more slanted towards the “not quite a real issue” category, or duplicates of the “issues” I didn’t fix.
By the third round of review, almost everything it produces will be useless.
I’m pretty sure this is what’s going on with Project Glasswing.
30 points on Google rejects UK request for Russell Brand YouTube video to be removed in !world@lemmy.world ·
Ahh. They don’t care about this video. They just want an excuse to censor people in the future.