Justin

2025’s flights! ✈️

My unwokest belief is that the dynamics of supply & demand are real and that no amount of wishcasting for a socialist utopia will change the economic realities created by restricted supply

@_sholtodouglas I doubt the actual real usage of those chats is anything meaningful, but it’s just a mind trick knowing it’s depleting it, whereas with Codex it’s entirely separated from chat.

@_sholtodouglas Not a model issue, but going to take the opportunity to share it anyway: the sharing of rate limits between Code/Cowork and Claude.ai chat. I prefer Claude but find myself reaching to GPT for chat queries knowing it won’t short change me later for real work.

When do you reach for other models instead of Claude? What can we do better? Hit me with all of your frustrations. dms open. If you can give me detail (e.g. specifics/transcipts) - it'll help a lot in finding out exactly what we need to do to improve the next model

@LinkofSunshine Way longer than 2 years

Disney bought 538 and then shut it down and burned it all within like 2 years…. What were they thinking?

@Phil_Lewis_ they said things would start getting weird...

Dozens of empty Waymos invaded an Atlanta neighborhood and circled a cul-de-sac for hours with no passengers wsbtv.com/news/local/atl…

@omgsidewalks there is no such thing as passive income, it's delayed income. you put in the work upfront typically underpaid or unpaid and it's 'passive' later only in the sense of slowly paying you for your previous work

Capitalists will really say things like, "So you just want money for doing nothing, is that it?" and then turn around and praise themselves for their passive income.

@UFOTofu42 @Aella_Girl @omgsidewalks no such thing as passive income, it's delayed income, you put in the work upfront and it's 'passive' later slowly paying for your previous work

@Aella_Girl @omgsidewalks ok but he's criticising hypocrisy, not passive income

The terminus has always been computers that use computers to make computers, no getting around it

@Duderichy every time I have a connection in an airport coming home from Japan I find myself instinctively still mini head bowing random people and getting weird looks like wtf are you doing

going to Japan has made me realize I always bow my head as a sign of respect in America

@hecubian_devil two years is frankly generous lol

Yes, this is slop—in part because the person making seems tasteless, dull, and infatuated with the tech itself rather than narrative art—but video gen is clearly close to *commercial viability* in the right hands. Especially for kids’ content, since they’re undiscerning. This is why I’ve been agitating for a leftist AI skeptical response bigger than “it’s slop, it sucks, it’s fake, it’s bad.” Because it has been getting vastly more capable, and is rapidly approaching the point at which *enough* people will be willing to pay for it—even if YOU don’t, even if LOTS OF US don’t—that it will entrench. Right now, we call it slop and get a lot of positive feedback and feel superior and good and like we are leading the majority. Give it two years and people will be parking their kids in front of AI-generated children’s shows without a second thought. They’ll be getting mad at you for calling it slop, because they’ll feel impugned as parents. Other people will be getting mad because they have AI “friends,” or they’ve gotten addicted to using it at work or school to take shortcuts. It feels so good right now to shit on AI on the basis of its capabilities, but that is clearly *time-limited* and ultimately a failing strategy to contain and regulate AI. we have got to come up with something better, stronger.

@deredleritt3r It’s actually still the prevailing view lol

Old enough to remember when the prevailing view on AI was that LLMs are a scam, actually, and the bubble is about to pop (6 months ago)

@CacheThatCheque debatable if we're a real country

I keep forgetting Canada is an actual country with its own stock market

@tinioril @mastaprincess @liminal_warmth literally

@mastaprincess @liminal_warmth Literally or are you referring to the ship?

I like how literally every article and tweet about this hantavirus situation is like “human transmission is possible BUT VERY RARE” and then cheerfully goes on to tell you a bunch of people infected each other in a short time frame on a cruise

@AaronBergman18 Uniqlo AirSense pants will save you

Fuck I think wearing pants instead of shorts actually does make you look nicer and more professional It’s over

@hecubian_devil Btw I completely agree with you that the concerns SHOULD be focused more on issues like surveillance that you point out

@hecubian_devil You have to remember too that the earliest displays of AI a few years ago that were pushed by companies revolved more around "look at this cute picture or story it can make for you!" than the more agent-like work we're seeing today. They haven't updated on the latter yet.

I am realizing I don’t understand what the Left is worried about with AI. I assumed we feared stuff (surveillance, propaganda, policing, social isolation, job losses) that hinged on the tech being economically viable—fears related to the AI models *working*, essentially. And to that end, I’ve tried to articulate what I see as the major concerns with overcommitment to the “it’s a scam” counter-narrative. Namely, if it’s a scam, if the tech doesn’t work, then at some point the bubble pops. People stop buying it and investors stop funding it, the companies go under, and the house of cards collapses. But in this scenario, most of the fears I *thought* we had, like about social isolation or job losses, disappear. AI becomes a mostly self-limiting problem that solves itself once it inevitably fails. My concern is that this narrative leads to passivity. If the AI scam will collapse on its own, you don’t need to actually take real action against it—not legislation or other state action, anyway. You can hasten the collapse, however, by spreading the truth, by piercing the bubble for more people around you. By posting, mainly. Obviously, this has a certain appeal. It locates the correct mode of action—consciousness-raising, basically—well in the hands of what individual people can do, immediately, by doing something they already enjoy (posting). It’s both personally easy *and* makes you feel like you have individual agency. But the more I talk to people about this (or the more I get yelled at, really), the more evident it becomes that people see the scam AS the threat. Which I don’t fully understand. I get that it’s a threat in the sense that a bubble popping can result in a financial crisis/recession, so it has the character of a short-term economic problem, but the rhetoric seems so much more intense than that. People talk about the *tech itself* in strongly moral terms. Evil, anti-human, perverse, brain-rotting, etc. That seems deeper and more fundamental than the threat of a transitory economic crisis. And I have trouble reconciling the idea that the bubble will pop and the tech will be exposed as fraudulent, yet the tech itself is *powerfully evil* and must be resisted—if it doesn’t *work,* how can the *tech itself* be such a threat? How can it be a threat if the companies making it go under, and people stop investing in it? When I ask people this, they sometimes respond with versions of “well obviously even when the bubble pops the tech will stick around, companies will consolidate, etc”—which I agree would happen in a bubble-popping scenario, but I also think the tech is economically viable because it can actually replace lots of kinds of human labor. I think it *works*, to some extent, so it makes sense to me that the tech would be persistent. But if it’s a *fraud* and a *scam*, I have trouble reconciling that. So I just end up being very confused about *where exactly people are locating the threat from AI*. From what I can tell, a lot of leftist anti-AI writers and experts seem to have real worries that the tech itself is—like many automation technologies before it—fundamentally economically viable, which will lead to social and economic dislocation as it gets entrenched. This is a perspective I understand. Whereas “the scam is the threat” feels like a more diffuse, popular perception. So, genuinely, I’m soliciting answers to the question: if you’re on the left and anti-AI, where do you see the threat coming from? How do you envision it manifesting? What do you think needs to be done to stop it?

@tekbog you can't hit them with both at the same time that's just dangerous

this is going to one shot my wife

@tomfgoodwin Probably need to specifically prompt against web searches for the first query, and then afterwards prompt to use web searches only to verify the previous suggestions actually exist and don't have hallucinations

I swear, the first time I used generative AI for recommendations, it was amazing I remember being in Kyoto and asking for a little cocktail place, and the answer was incredible , I gave a very vague prompt and it "knew me" I did the same in Melbourne and with remarkably little input, it found me 3 stunning unexpected places. I asked for atypical things in Miami, And it suggested places that I'd never heard of But now, if you do it, it basically just uses TripAdvisor and Reddit, or Google Maps. If you asked for places to go in NYC it will tell you the High Line, not like the past when it could say OCDChinatown It went from being magical, astonishing, and invaluable, to rather useless. What happened? Commercial deals ? Laziness?

@Romy_Holland In that vein, I recommend spending multiple days in Kyoto, going to one of the TeamLabs, day trips to easily accessible places from Tokyo/Osaka like Kanazawa, Hakone, Nikko, or Miyajima, and even going to Tokyo Disneyland if you like theme parks. Have been to Japan 7 times.

@Romy_Holland I think you should trust that many things that appear as overly recommended tourist activities are in fact actually good and worth doing, don't dismiss them just because they're popular. At least in Japan, they are often very good and popular for a reason.

i want to solicit japan recs but i feel like the vast majority of people just enthusiastically recommend the things they happened to do in a given place, but they actually have no idea if these are the best things to do. so i guess im specifically requesting japan recs from anyone who has reason to believe they know the actual best stuff to do. like what would you fly back here to do again? what have you found yourself thinking about for years after the fact?