yreg

5 ms·

1.
by yreg 4d ago
For me it's annoying that I cannot right-click the link and copy the actual target. They are breaking the web.
2.
by yreg 5d ago
These things were obviously always concerning but I feel the situation has taken on an entirely different dimension now that they could feasibly use whisper + LLM to actually review _all_ the communication. That's crazy.
3.
by yreg 7d ago
Darkness travels even faster than light on Discworld, because it has to be fast enough to get out of the way when light arrives.
4.
by yreg 7d ago
It is a bit less strict than you think. It's not just about the police giving the tourists some leeway. - Alcohol is fully legal for non-muslims, even if they have residency (or in rare cases citizenship). - Unmarried cohabitation is f
5.
by yreg 8d ago
So this is explicitly built for partially-contained agents who have the normal protocols/methods blocked, right?
6.
by yreg 8d ago
Wasn't holocaust caused by words?
7.
by yreg 8d ago
You don't need to cancel all the noise to solve 1+1 correctly. The model in the example is non-deterministic. I don't know Penrose's view deeply enough to defend it further (view which I for the record do not share).
8.
by yreg 8d ago
For the hardcore Overwatch community perhaps, but not for the general gaming community, especially if WG is about to grow further and host more and more wikis. For people who regularly get into new games its much easier to learn just one ea
9.
by yreg 8d ago
Per the article they don't. > There’s a bit of light at the end of the tunnel, though - based on our experiments so far, it seems like once we establish these new wikis fairly well on Google, it should be safe to move them off weird
10.
by yreg 8d ago
Fair enough, I can see that.
11.
by yreg 8d ago
If you run an LLM (with temp >0) and use a true random generator for seeds, it will still tell you 1+1 is 2.
12.
by yreg 8d ago
I'm with you regarding opinion on consciousness, but do you think it would be 'better' if there were no dualists and no dualist essays and ideas? Because I hardly think so. This is not like alternative medicine where there is
13.
by yreg 8d ago
> If mind was random, people would miscalculate 1+1 with 50% probability. That does not follow at all.
14.
by yreg 8d ago
Thank you so much for what you do. > Of course the alternative is what the parent comment suggests ("probably just need a good base domain for game wikis") - but you have to get THAT domain out of Google jail first. So why not
15.
by yreg 8d ago
Have you actually hear him talk about it? It is very far from rambling and he will be the first to tell you that he is not sure he is right. We need physicists to come up with and develop novel ideas to consider no matter how many of them m
16.
by yreg 9d ago
Don't plenty of android devices have barometers? I expect Google being Google and recording all that data unless you opt out. Would that be practical for weather forecasting or not really?
17.
by yreg 9d ago
It is odd but also fascinating. If he likes it that way then all the power to him. He's not exactly trying to convince others to do the same.
18.
by yreg 9d ago
Were people not happy about WSL?
19.
by yreg 9d ago
I too liked the detour about cheating.
20.
by yreg 9d ago
Why sell? I thought the idea of donating to them is to make the information available for everyone for free.
21.
by yreg 12d ago
So he is posting it again because he believes it is crucial information for us. Apparently the crowd agrees because they keep upvoting these.
22.
by yreg 12d ago
I don't think it's truism. One could believe (and I think plenty people do) that with the right containment it is possible to contain an intelligence regardless of its strength.
23.
by yreg 12d ago
I still don't get how do you do this undercover. There are a billion websites. How do the agents stumble upon the same obscure unused wikis?
24.
by yreg 12d ago
Maybe I'm leaning into scifi, but I believe that Yudkowsky is right that a sufficiently smart intelligence is uncontainable at all. We can only hope to either never create an AI so strong or to align it correctly. But if it is not alig
25.
by yreg 12d ago
I don't understand one step: How did the agents know to gather on that particular website? Did the second agent just google for something like it and find the first one's post?
26.
by yreg 13d ago
What's so dangerous about it?
27.
by yreg 13d ago
Oh, yeah. That happens frequently as well.
28.
by yreg 13d ago
Why are you so cryptic? Just say what you mean plainly.
29.
by yreg 14d ago
It's pretty annoying when traveling to new countries, which I do often. Always figure out all the apps I need to use local transport, to park, to charge the car, etc. In plenty of regions you also need special (often confusing) wallet
30.
by yreg 14d ago
Then why choose the most inconsequential example (IMO) I can imagine? And what's the broader issue? That it's moderated?

More ›