I don’t want the US to win anything ever again. They are a net negative in this world, obsessed with short term profits. Countries like China with long term objectives are better.
I'm a bit sad that the US now look less reliable (and, on average, more of a net negative) than China, but I can't blame anyone who has that impression right now. As long as the guy who sets the long term objectives is reasonably sane, I guess? Putin attacked Ukraine, Trump (and Bibi) attacked Iran, I'm hoping Xi is smarter and doesn't attack Taiwan, otherwise we can all say good bye to our jobs, and who "wins" AI will be the least important question on our minds...
It’s all about appearances isn’t it. America has dropped more bombs than China. When did China bomb a country last time? But appearances change and that is sad.
The only thing that changed was that America turned on its client states and started saying unhinged shit instead of appearances-speeches like spreading freedom and democracy.
Edit: replaced “has bombed more countries than China” with “dropped more bombs”
America’s goal is to rule the world through a hegemony built on fear and violence. With the current administration, it’s just that the mask has slipped.
Americans are all the same really, it doesn’t matter who voted for who: FIX your shit, fix your society. Stop oppressing the world with destructive capitalism. And yea maybe that means you have to slow down and live a little more inconveniently, but it will be for the betterment of the world.
I think China has too much to gain by looking like the word's stable alternative to America. What really gets me depressed is their current trajectory towards cultural homogeneity, even internally. There seems to be a real push to calcify the Han identity. I don't know why dictators are so drawn to homogenizing their own cultures when nature itself seems to point to diversity as strength. It leaves me still kind of hoping America is somehow able to pull its head out of its ass, but they'll have to overcome their own oligarchy and entrenched structures - Trump was a symptom.
China took a very rich business man and told him to stop showing his richness and start doing more for china.
China has a real plan for renewable energy and pushing through it.
China is smarter because it doesn't allow some people to vote for people like Trump and its smarter than russia because it is less motivated by one persons personal agenda.
you don't really need to explain that to someone with common sense. If someone says a communist countrie should be the top of the world, they are obviously part of the communist party or simply re*rded.
I am puzzled by this sentence, which combines nationality, psychosis, and intelligence into one. What if the parent commenter is Vietnamese? Or Hungarian? Or Turkish? Will this fall into the "or" clause?
Beside posturing, modern China was only involved in skirmishes at the Indian border. What's delusional is blindly comparing China to Russia or the USA.
From company metrics I have found that developers who make a lot of mouse movements correlate with weaker performance reviews. Something to think about.
"What it means to be human is to work 16 hours a day for someone else taking home just enough salary to survive another day, because you too, someday could become a millionaire" --HN
A lot of what humanity does seems to be a persistent terrible take.
I hope you count "stimulating our minds for either learning or imaginative purposes" as one of those outcomes because if you only count "work produced and kpis met" as an outcome then that sounds pretty bleak.
I think at best we will find very distant evidence of advanced alien life: Inspiring, but so far away and unreachable that we’ll never meet or communicate with them in any way.
Don’t know what’s worse: wondering if there is alien life or knowing there is alien life in a distant system somewhere but having no further information that can be learned about them.
Or the implication that FTL and interstellar travel is either outright impossible or completely impractical. We are trapped alone in our otherwise unexceptional solar system forever, or until there is a catastrophe so large as to kill us off as a species.
I kind of feel like in the old days, people weren’t really afraid of heights. Heights were fairly new, and exciting, and the consequences of falling were not well understood yet. It’s why you would see skyscraper construction workers jumping around and sitting on beams to enjoy a casual lunch thousands of feet up, without a care in the world.
…the consequences of falling were not well understood yet
Eh? Cliffs, trees; fall off anything at 10 meters or higher, and your odds of dying are around 50%. I’m pretty sure folks were aware of the consequences long before Eiffel showed up.
Perhaps it was more that risk taking and dying young from various illnesses was a thing that happened around you a lot, so you were more accustomed to it and didn't expect to live to 80.
If I was a professor, I would make a very clear policy they AI is not to be used on assignments, and would repeat it throughout the semester, but make no effort to actually enforce this and even make it easy to abuse.
Then, for the final exam, drop the bomb: in person, handwritten, no outside references, mostly the same assignments we’ve done before. If you fail, it’s over for you. If you stayed true and studied, it should be easy for you to pass. If you used AI all semester, you did it to yourself. Those who complain will have their past assignments audited and if AI was used they are reported for plagiarism. This will be the most valuable lesson.
> It will never be cheaper than what it is today. Anthropic is heavily subsidizing.
We don't know that for sure—they've dropped prices before:
1. Claude 3 → Claude 3.5/3.7 generation (mid-2024 to early 2025): Haiku went from $0.25/$1.25 to $0.80/$4.00 per MTok — this was actually a price increase for Haiku, but Sonnet stayed flat at $3/$15 while delivering significantly better performance, effectively a price-per-capability reduction.
2. Claude 3/4 Opus → Claude Opus 4.5/4.6 (late 2025): This was the big one. Opus dropped from $15/$75 per MTok down to $5/$25 per MTok — a 67% reduction on input and output. This is the most significant explicit price cut Anthropic has made, delivering a far more capable model at one-third the price.
They're definitely not subsidizing API pricing, can't believe how prevalent that fallacy is on HN of all places. The question is how profitable Claude Code is. Your example 2 is real and major but your example 1 is ridiculous, almost any new model from any company is better at the same price, and how is increasing the price an example of decreasing prices??
BTW, Github Copilot is pricing Opus 4.7 at 2.5x the cost of Opus 4.6 at promotional pricing (so maybe it'll be 4-5x). But Github's request based pricing is insane, completely divorced from their actual costs (you can achieve 1+M tokens for $0.10 if you give it a large request), so I'd assume they're losing a lot of money.
The cost of a thing, is relative to its source costs. They are subsidizing API pricing, if you consider all the costs to provide the service, including all model creation, training, etc costs.
But that doesn't mean they will be more expensive, longer term. The cost of compute will go down as time goes on. Each year it will get cheaper. Same for power requirements, computing density, cooling, and so on.
I remember trying to store and play mp3 files on older computers. I could typically hold a few on a disk, and if I wasn't doing anything else I could play one. Barely. Now you'll be hard pressed to play an mp3 and see the load results in top or what not.
If those cost of compute is going down, then eventually it will go down enough that we will run on our LLMs locally and Anthropic will go out of business.
> then eventually it will go down enough that we will run on our LLMs locally and Anthropic will go out of business.
I want robust local LLMs as much as the next person—Gemma E2B, 3.2GB does my word completions as I type. It's gotten to the point where it knows what I'm going to type before I do!
But I don't see Anthropic going out of business anytime soon. As good as some of the open source LLMs are, we’re still a long way from being able to frontier models at home.
If you are using LLMs for tool use locally, then in a decade it will not make sense anymore to pay for hosted solutions. Your device will have compute power to run powerful LLMs trivially.
If you need LLMs at scale to serve many customers, then hosted solutions make sense for the availability aspect. But by this point models can be offered by any generic services provider, like AWS or Cloudflare. Pure AI companies that just offer hosted models and nothing else will go extinct if they don’t expand to offer more services.
> If you are using LLMs for tool use locally, then in a decade it will not make sense anymore to pay for hosted solutions. Your device will have compute power to run powerful LLMs trivially.
LLMs a couple of years ago that'd be impossible to run on consumer hardware are now running on consumer hardware. I'm less concerned about compute power; it's more about memory.
It could be several years before new RAM capacity comes online. Even then, it won't be cheap.
I expect in the future, hosted frontier models will be a utility like electricity or cable tv. Part of a package most people will subscribe to.
> can't believe how prevalent that fallacy is on HN of all places
AI is very emotional for a lot of people leading to bias takes in both directions. We like to think HN is more rational than average, but we’re all human.
It's partly that, but also reading and surface level understanding something vs generating yourself are different skills with different depths. If you're learning a language, you can get good at listening without getting good at speaking for example.
There should just be a way to setup an alternate dummy account based on the finger you use. This gives the illusion of compliance but your real data is safe.
If you're in a situation where this is a pressing issue, it's not a good solution as it's trivial to detect if it's a fake environment, especially if they get suspicious and run external forensics on it.
iirc the GrapheneOS team won't implement this feature for that reason
reply