There is also the problem that making platforms responsible for policing user-generated content 1) gives them unwanted political power and 2) creates immense barriers to entry in the field, which is also very undesireable.
I have no idea how to define it. I also don’t know if I’m personally convinced one way or another about the harms. Just think the platforms would probably have to be made to make more substantial changes were it the case.
> Rahm Emanuel is campaigning for the same thing in America.
I get the sense this is supposed to signify something; don't know the name, but looking at their profile, great career, Obama's chief of staff. What's the implication?
He is a pro-authoritian-control Democrat, so it is unsurprising that he is more worried about control of information than he is the Constitution. His background is in finance and his political goal is generally management of the country by a monied elite without particular oversight.
He was paid by Goldman Sachs to help Clinton get elected by raising massive amounts of money. During Obama's term he structured the DNC to be about his personal power rather than supporting Democrats across the country, costing Democrats the midterms. As mayor of Chicago he covered up a murder committed by a police officer and refused to comply with transparency laws.
On the other hand, this particular position is probably just part of the Israeli campaign against TikTok: Emanuel volunteered for the IDF and has long been an anti-Palestinian activist.
LLMs are “concept based” too, if you can call statistical patterns that. In a multi-modal model the embeddings for text, image and audio exist in the same high-dimensional space.
We don’t seem to have any clue if this is how our brain works, yet.
We know empirically that time only flows in one direction, it can’t be described as just a perception. You’d have had at least the tiniest of evidence that time sometimes flows backwards.
Time doesn't flow at a speed. So, time flows at no speed, so, time doesn't flow. Time doesn't exist within time, so it has to be static. Moments don't change.
How would you measure time going backwards if you can only perceive it going forwards? How can you "experience" everything around you going "backwards" if that includes your memory? How can you determine that a specific moment in time was arrived at by time going forward, or by going backwards?
How do you know anything outside of your perception is true? All things boil down to a philosophical argument. The simplest answer is that "time" as we imagine it is a product of our interpretation and the true nature of "it" is hidden from us.
This is about time as it relates to our understanding of the physical world through science. You might as well say matter is an illusion, nether is relevant to this discussion.
You believe it is irrelevant - I believe it is relevant. I argue matter itself has more basis in reality in that it cannot be as easily explained away as time. In the context you're speaking in, there is much more evidence and logic to support the existence of matter. My point is that time as we believe it exists, is a construct that has meaning to us because of its benefit to our survival, rather than it being an objective reality.
_Everything_ flows in one direction, all particles goes in a straight line from their self reference, fields "modifying direction" is just an observer point of view. The separation of time and space is purely a perception matter.
A gross comparison would be to compare with objects perception, it only exists because our mind can leverage it for a strong evolutional advantage (I'm not only speaking of humans here).
No, a particle can flow left OR right, up AND THEN down, forward THEN reverse THEN forward again.
But in time, it can only go forwards, at very slow rates like far from gravity wells, or fast like in relativistic situations. Never backwards. Never stop.
As we haven't seen any evidence of this, then the effects must be so tiny that we can just ignore that possibility. It's like worrying about the gravitational effects of Russell's Teapot.
One of the reported hallucinations in this work [1], starting with David Rein, says the other authors are entirely made up. They are indeed absent from the original cited paper [2], but a Google search shows some of the same names featured in citations from other papers [3] [4].
Most of the names in these wrong attributions are actual people though, not hallucinations. What is going on? Is this a case of AI-powered citation management creating some weird feedback loop?
I noticed that for the last few years, my recommendations are almost the same for every category. 'Top picks' is 60% the same as 'We think you'll love this' which is almost the same as 'Your Next Watch' which is similar to 'Award-winning movies', 'Chilly thrillers', repeat ad-infinitum. If there is a new Netflix-owned movie it will definitely be in there.
Then on top of that, similar to YouTube, half of that content are things I have already watched. HBO and Amazon are even worse in this aspect but it just drives me crazy, feels like seeing the same 100 movie options over and over for months. Has the catalog shrinked that much over the years?
I started keeping a separate list of films to watch on IMDB, but 6/10 times they are not available on any service except for rent in AppleTV.
That's kind of what they are doing - the move is 'updating their CI' to Codeberg Actions which is presumably more reliable. All the git workflows stay the same.
There is also the problem that making platforms responsible for policing user-generated content 1) gives them unwanted political power and 2) creates immense barriers to entry in the field, which is also very undesireable.
reply