BosunoB

No, it has knowledge of what it is and how it is used.

I'm guessing you and the other guy are taking issue with the words "aware of" when I'm just saying it has knowledge of these things. Awareness doesn't have to imply a continual conscious state.

I didn't mean to imply that it was. But when you reply to it, if you just say "no" then it's aware that you could've just not responded, and that normally you would never respond to it unless you were asking for something more.

It just doesn't make any sense to respond no in this situation, and so it confuses the LLM and so it looks for more context.

The whole idea of just sending "no" to an LLM without additional context is kind of silly. It's smart enough to know that if you just didn't want it to proceed, you would just not respond to it.

The fact that you responded to it tells it that it should do something, and so it looks for additional context (for the build mode change) to decide what to do.

Yes, because current unemployment comes as the result of complex factors that intersect with various groups and ideologies in complex ways. Also, raising employment is a complex task.

When AI takes the jobs, it will be dead simple to the majority of people that the current way of doing things will not work, in the same way it's dead simple to you and me.

Y'all gotta stop looking at politics this way.

You know why they don't share the fruits of capital with us now? Because Americans hate getting taxed to pay for welfare, and so they've been voting against taxes for 50 years. This whole political landscape changes when people lose their jobs to AI, a thing that everyone thinks should be taxed. In fact, the entire ideological underpinning behind extreme wealth accumulation is gone when AI runs everything.

I was able to find the podcast. Question is at 33:30. He doesn't give hard data but he explains his reasoning.

https://youtu.be/mYDSSRS-B5U

Dario said this in a podcast somewhere. The models themselves have so far been profitable if you look at their lifetime costs and revenue. Annual profitability just isn't a very good lens for AI companies because costs all land in one year and the revenue all comes in the next. Prolific AI haters like Ed Zitron make this mistake all the time.

There are companies working on this, but my understanding is that the training data is more challenging to get because it involves reinforcement learning in physical space.

In the forecast of the AI-2027 guys, robotics come after they've already created superintelligent AI, largely just because it's easier to create the relevant data for thinking than for moving in physical space.

Robotics will come in the next few years. If you believe the AI2027 guys, though, the majority of work will be automated in the next 10 years, which seems more and more plausible to me every day.

Most of that 9 billion was spent on training new models and on staff. If they stopped spending money on R&D, they would already be profitable.

I switched to Floorp last month and it's amazing. It's a Firefox fork with tab rows. Tab rows are the real MVP. None of the annoying weirdness of Tree Style Tabs, where you have to keep track of a hierarchy of tabs that are hidden behind other tabs. Instead, you just see 3-4 rows of tabs and you make a mental map of what is where.

Once they release the new version in a month or two, we'll also get newer Firefox features like these tab groups, and we'll also get workspace improvements. Floorp is 10/10.

You are absolutely right that AGI will probably barely resemble LLMs, but this is kind of beside the point. An LLM just has to get good enough to automate sufficiently complicated coding tasks, like those of coding new AI experiments. From there, researchers can spin off new experiments rapidly and make further improvements. An AGI will likely have vastly different architecture from an LLM, but we will only discover that through likely hundreds of thousands of experiments with incremental improvements.

This is the ai-2027.com argument. LLMs only really have to get good enough at coding (and then researching), and it's singularity time.

I fell into Rand in high school and it took me a few years to climb out.

The problem with believing in the primacy of reason is that it's incredibly distortionary. In reality, we all think and reason with respect to our ego and our emotions, and so if you believe that you are engaging in pure reason, it can lead you to pave over the ways in which your emotions are affecting your line of thought.

In this way it can quickly become a very dogmatic, self-reinforcing way of thinking. The ironic thing is that becoming a better thinker is not done by studying logic, but instead by learning to recognize and respect your own emotional responses.

My advice, which you might not see here, is to write about every game you play. And ideally every movie you watch and book you read, too. To develop a sense of your own taste and what makes a game good to you, you have to write your ideas down. Go through the major moments of a game and describe how they made you feel, and what they did to make you feel that way. Dissect the mechanics and identify stuff you liked and stuff that frustrated you.

This is pretty much what college is. You just write essays where you dissect stuff like that and then a professor reads them and gives feedback on your ideas and how you communicate them. You can do the same thing informally with your friends and partners, just discuss games.

For anybody looking to try psychedelics without the sketchiness of finding a dealer or buying online, you should know that you can discretely grow psilocybin mushrooms in your home for less than a hundred dollars initial investment, and with very little effort. It's legal to buy the spores in most states, and it only takes a couple of months to grow a nice canopy.

The best guide is https://www.reddit.com/r/unclebens/ or https://www.reddit.com/r/shrooms/comments/8e7g6n/how_to_grow.... Expect to read a lot before you start growing.

This might be true if Spotify's monetization was distributed fairly, but instead it's basically redistributive to the biggest players. In short, instead of splitting a person's premium payment equally among all the artists they listened to, it's split among all artists on the platform. E.g. if Lady Gaga is 2% of all Spotify plays in a month, they get 2% of the money, even if you personally didn't listen to Lady Gaga at all.

It's the paradox of tolerance. If you tolerate the intolerant (e.g. right-wing assholes) they will push out other groups through their intolerant behavior. The only solution is to rabidly ban hate speech and similar behavior.

I'm pretty sure 90% of the changes won't impact anyone's lifestyle. Moving to renewable energy and phasing out ICE cars allow for the exact same lifestyle. The only things that might have to change are things that require carbon emissions, like plane trips or meat production (although this one also might not require carbon emissions in 20 years). I guess you could argue that slightly more expensive manufactured goods impact a person's lifestyle, but that's a pretty minor change, and if you do a carbon tax and dividend, that's going to be moot for all but maybe 10% of the population.

For anybody interested in this, you should also check out Path of Diablo https://pathofdiablo.com/. It's a popular mod with a lot of QoL improvements and which borrows Path of Exile's robust endgame mapping system for replayability.

>Having "UBI" in US would create a skyrocket inflation because worker who are paid 10$/Hour would drop their work and stay at home and do something that is more profitable for them.

Unless you have some hard data, your argument doesn't seem very convincing. The numbers could easily work out in any way, and it would probably depend on the actual amount of UBI you use, because the effect on the workforce and the spending power of consumers would be important for determining the amount of inflation. Studies are useful in situations like this.

Also, I mean, this subtly doubles as an argument against a minimum wage increase. In the same sense that a UBI might slightly decrease the workforce and lead to higher wages for those remaining (and inflation), a minimum wage increase would also. Do you have a proposed tool to increase the wages of an immiserated 20% of the population that doesn't also have the potential to cause inflation?

I think that people can accept a universal program, but legislators and the think tanks that govern their policy cannot. The point of Dorsey's funding is to provide high quality data to bolster the policy, and this might have that effect even if, like you say, the payments aren't universal.

Sure, I wasn't trying to argue against a green new deal or anything, just to give a justification for a UBI in a way others haven't. Clearly they aren't mutually exclusive, unless you're the sort of person who thinks a UBI will mean nobody works any more.

Renters markets suck but it's not like markets just cease to function as soon as consumers get more income. I'm sure there would be some rent seeking, but I doubt most renters markets are so uncompetitive that they could gobble up 1k or whatever.

Which European countries have a basic income?

Other people have offered practical reasons why they aren't really the same programs with the same goals, but in case you're up for getting utopian, some people see UBI as a necessary step as we enter a later stage of capitalism. UBI + reduction of working hours together are two feasible steps to reevaluate life under capitalism. Rather than sticking people in a 40hr/week job in order to survive, these steps give people more agency over their lives and probably improved treatment as workers. Basically you're paying people to live their best lives and breaking down as many barriers as possible for them to do that.

Hot take: 95% of major releases are bad. So if we get up to 20 video game movies and there still isn't a good one, then maybe we can let ourselves think there's a problem.

I worked on a small project in Godot and I found the transition from Python to be pretty natural. Here's the GDscript primer if you want to take a glance:

http://docs.godotengine.org/en/3.0/getting_started/scripting...

This isn't a completely sufficient argument, though. Surely, any person into an artform will at some point be able to recognize the difference between what is good because it is good and what is good because it is different. A good critic would be able to differentiate these two things and give a review with some nuance, but it seems incomplete to say that the critic problem is mostly the fault of seeing way too much junk. Anybody who is actually in to anything has seen a ton of junk, but not everybody has lost their ability to clearly judge things.