Hacker News Reader

Comment by data-ottawa | Hacker News Reader

As a non-US citizen I guess this is the last money I pay to US companies for AI then.

I can't help but wonder if it's now obvious that frontier AI work should not happen in the US.

I can understand the KYC aspect of this, but at the same time, how can anyone trust US based AI after this? Maybe this is a continuation of the Pentagon feud, or it's revenge, or it's a KYC play. Either way, you've got a government willing to shut down companies sales over arbitrary reasons.

Ironically, I mostly have a subscription to Claude for work, which is primarily for US baed companies.

View on HN

In my experience, US citizens are completely blind to how much stuff like this makes citizens of other countries hate their government (which often, unfortunately, bleeds over to hate for US citizens; not that I condone hating any group of people based on the actions of their state)

The US has spent the last 12ish years betting that they're the only country that matters, but the end of result of that is that somehow when I talk to Australians in my age group the average person has a more positive opinion towards China even than the US.

The default orientation of Americans toward government is already skepticism and distrust. The average person is questioning "why did you ever like the government in the first place?"

I don't believe that at all, the average person voted for this government.

The average person didn't vote.

If you're referring to voters staying home, staying home is a vote for their state's majority. There's no sense in which the average citizen in a democracy is not responsible for the outcome of a democratic vote.

That’s not even factually true. Turnout in 2024 was 63.7%[1] of the voting age population. You’re just wrong.

[1] https://ballotpedia.org/Election_results,_2024:_Analysis_of_...

If you include people not allowed to vote (e.g under 18) it is below 50%

What do you think "voting age" means? The first sentence on the linked page is "The overall turnout of eligible voters in the 2024 general election was 63.7%"

Huh, I must have inverted that number in my head. Thanks!

This is categorically wrong, but even then - those that do not vote at all bear even more responsibility, and are traitors to democracy itself. By not participating in shaping it, you're dishonouring everyone that fought and died for your freedom.

Not voting is a vote for the status quo.

Not being politically active is a political choice itself.

Is that any surprise? China has been very good about not fucking with other countries even though they absolutely have the capability to.

I think that really depends on which country you live in. My country has only had a few relatively minor spats with China, Vietnam less so.

Still much less that the US.

The average American voter primarily uses their vote in an effort to hurt other people who might support a different team.

> which often, unfortunately, bleeds over to hate for US citizens

err... you guys voted for the administration that fucks over the rest of the world. Twice. So tell me, why would you expect the rest of humanity to show any kindness to the populace entirely responsible for what is happening right now?

You do know not all of us voted for him, not even a majority. You could argue the non-voters allowed this outcome. Should we discard the whole democracy thing because we don't like a result?

I lament that there wasn't a stronger candidate running against him, but the Democrats didn't have a primary, and even if they did, I'm an independent and do not vote in primaries ( this has changed in Colorado thankfully). A different, stronger candidate could have likely beaten Trump

Democracy means that you bear collective responsibility for the government you voted for, regardless of the choices of the individual.

I disagree and don’t see where you get that from. Why would that be the case?

I agree with you that restricting access to Fable is stupid, but I'm in favor of e.g. GPU export controls. It's certainly annoying, but—well, I don't know where you live, but you don't want to make it easier for China and Russia to build weapons they can use attack to attack Taiwan and Ukraine, right?

And the nice thing about the GPU restrictions is that even if they don't work completely, just making the hardware more difficult and expensive to access is useful.

I'm a US citizen and though I don't think you should hate people due to their government, I do think you can attribute some responsibility for the government actions to the people.

If we live in a democracy, then we are responsible for the actions of our government.

You do not live in a democracy or anything close to it. It's a constitutional republic, but the representatives are bought and paid for by people who do not have the public's best interest in mind. I do not accept responsibility for things I not only have no control over whatsoever, are governed and controlled by forces much more powerful than I could ever realistically be, even if I wanted to.

It's frankly a delusion many americans still hold that they can somehow vote their way out of this situation. I don't know how at all it is salvageable, I think best case for Americans and probably the world is that we slowly lose all relevance until the powers that be get bored and go prey elsewhere.

So you're going to use DeepSeek, Qwen, GLM, Kimi and Mistral now? I tried them, and they really fall short of GPT and Claude.

Without access to US models, I'd be limited to asking simple questions in chat interfaces and maybe some grunt work in coding CLIs, but even that the weak models will mess up.

Nothing has reached Opus and GPT5 levels in my personal experience, which also aligns with what the labs themselves admit ("near-frontier").

Well I am definitely not using the models that I'm not able to access.

So now the question is whether the capabilities of other models are worth their far cheaper token prices.

Plus, are we at all confident Opus or GPT 5.5 aren't about to get shut off?

Not all people need the SOTA. Also, many take into consideration speed, token / plan cost and many other factors when choosing a model

> Nothing has reached Opus and GPT5 levels in my personal experience

You mean, GPT 5.5 xhigh and Claude Opus 4.8 max? At least the benchmarks / public evals / rankings show some of the new coding models (ex: Qwen 3.7 Max & Mimo v2.5 Pro) are Opus 4.7 & GPT 5.4 level (but 3x to 5x cheaper): https://artificialanalysis.ai/leaderboards/models / https://gertlabs.com/rankings Personally speaking, in the past 1mo or so, I haven't missed GPT 5.4 / Opus 4.7 after moving to Qwen 3.7 / MiMo 2.5 / Kimi 2.6 et al.

That is very promising news. I will re-eval them all shortly. And you are suggesting that a higher reasoning budget can make up for weaker per-token performance? That is indeed worth evaluating.

Comparisons using the vendor-specific effort is apples and oranges. Ideally the evals would use a thinking token cap or something, so we can compare per-token performance. But eval is hard enough as it is.

I have been using DeepSeek at home. I have access to Claude and ChatGPT at work.

I honestly think that DeepSeek is as good, and sometimes even better, than the competition.

Reply on news.ycombinator.com