Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

There is minimal downside to switching to open models

marble.onl

Post reply on HN
Print view View original

221–230 of 351 posts

Re: There is minimal downside to switching to open models

#221

Post by vrganj »

cpursley wrote: post #112
lmm wrote: post #107

Earlier quoted context omitted.

> With all the issues in the US and generally wrong direction, I can’t remember them ever arresting people for mean tweets in the way that Germany and the UK have. Then you haven't been paying attention. The constitution prevents citizens from being convicted, but that doesn't stop arrests or being turned away at the border (even for permanent residents who've lived in the US for decades), and US citizens don't seem…

[flagged]

Why does it need to be a US citizen? Is mistreatment okay if they're not a citizen? That clause reads extremely chauvinist to me.

Re: There is minimal downside to switching to open models

#222

Post by vrganj »

nxm wrote: post #168
leptons wrote: post #121

Earlier quoted context omitted.

One guy spent 37 days in jail for re-posting a thing trump said ("We have to get over it" in reference to a school shooting), after Charlie Kirk was killed. https://www.bbc.com/news/articles/cg7pyjxjxrvo

So one guy, and now he’s suing which is a form of justice. No such path in the UK…. You’re fined and/or arrested and that’s all

Are you trying to say one can't sue in the UK? I'm confused by why you would think such a thing.

Re: There is minimal downside to switching to open models

#223

Post by senordevnyc »

bonesss wrote: post #50
realusername wrote: post #39

Earlier quoted context omitted.

There's also a lot of benchmark trickery going on, it's becoming harder to see how the latest models really improved. The top models also seem to have inconsistent performance depending on the time of day and how far we are from the next release.

I’m an LLM fan, but from an engineering perspective the idea of building atop services that palpably fluctuate in capacity, performance, and capability is nutty . Even with minor automation I feel like I can watch OpenAI and Anthropic engineers fiddling in real-time. Tuesdays behaviour changes by Thursday, 10AMs production isn’t possible at 11:30AM. Nutty.

Imagine having a business where you're at the mercy of the fluctuations in capacity, performance, and capability that your human employees display!

Re: There is minimal downside to switching to open models

#224

Post by fg137 »

coffinbirth wrote: post #70

> Open models are served via various means, some by the companies that released them and some by third parties like OpenRouter. Unfortunately, both of these routes are dodgier in terms of privacy and data sharing, and I would not feel the same comfort sending API calls containing client or confidential data to them. That's why I'm using eurouter.ai with the following routing rule for all my requests: { "model": "glm-…

If your only concern is data residency, data privacy and sharing, why not just use bedrock with the processing region locked to eu-west-2? For sure, it's not an European company serving the LLM, but it satisfies your requirements otherwise and is trusted by tons of companies worldwide.

Re: There is minimal downside to switching to open models

#225

Post by pell »

NamlchakKhandro wrote: post #207
leptons wrote: post #119

Earlier quoted context omitted.

>and US citizens don't seem to care I think maybe you haven't been paying attention. Most of us do care. Trump's approval rating is pretty low at 36%, and his disapproval rating is high. Just because he's still causing chaos doesn't mean the majority of us don't care about it. There's just no legal way to remove him, and his cronies simply won't do it - there's not enough votes in congress or he would have been gone…

NPR as your source isn't exactly a compelling argument tbh

Whenever I ask people to explain their issues with NPR it’s some cherry-picked news articles here and there that were somewhat biased. In my experience NPR often tries to be incredibly neutral, almost comically so, when criticizing any administration.

Re: There is minimal downside to switching to open models

#226

Post by quikoa »

fg137 wrote: post #224
coffinbirth wrote: post #70

> Open models are served via various means, some by the companies that released them and some by third parties like OpenRouter. Unfortunately, both of these routes are dodgier in terms of privacy and data sharing, and I would not feel the same comfort sending API calls containing client or confidential data to them. That's why I'm using eurouter.ai with the following routing rule for all my requests: { "model": "glm-…

If your only concern is data residency, data privacy and sharing, why not just use bedrock with the processing region locked to eu-west-2? For sure, it's not an European company serving the LLM, but it satisfies your requirements otherwise and is trusted by tons of companies worldwide.

> it's not an European company serving the LLM

That's a pretty big downside if data privacy and sharing is one of the main concerns.

Re: There is minimal downside to switching to open models

#227

Post by tsss »

nessex wrote: post #88
LPisGood wrote: post #73

Earlier quoted context omitted.

People talk about this a lot. What I have never seen is a discussion of methods they might employ to degrade the models. Let’s say I’m a bad faith LLM operator, and I want to degrade my model so the next release looks better and people want to switch to the more expensive one. How would I do that?

They would quantize the model. That'd make it cheaper to run, and have slightly worse output but it would still generate outputs with a similar feel, derived from a compressed version of the same knowledge base etc. They wouldn't even need to do this uniformly, quantized versions of the model could be routed only a subset of the requests. They could do this to nerf the old model, or more likely just to give themselve…

And guess what all the providers of open models do: They quantize, badly.

Re: There is minimal downside to switching to open models

#228

Post by helloplanets »

fg137 wrote: post #224
coffinbirth wrote: post #70

> Open models are served via various means, some by the companies that released them and some by third parties like OpenRouter. Unfortunately, both of these routes are dodgier in terms of privacy and data sharing, and I would not feel the same comfort sending API calls containing client or confidential data to them. That's why I'm using eurouter.ai with the following routing rule for all my requests: { "model": "glm-…

If your only concern is data residency, data privacy and sharing, why not just use bedrock with the processing region locked to eu-west-2? For sure, it's not an European company serving the LLM, but it satisfies your requirements otherwise and is trusted by tons of companies worldwide.

Anthropic already explicitly communicated that they'll store and check all the data from Bedrock or any platform, even if you've selected zero data retention, if using Mythos class models. To use these models on any platform, you'll have to accept these terms regardless of the region.

> Limited data retention and review as part of our safety work. Prompts submitted to, and outputs generated by, Mythos-class models are retained for 30 days for trust and safety purposes, on every platform where these models are offered.

> Change applies to organizations that have set up workspaces with zero data retention (ZDR) in Claude Console, use Claude Code with ZDR in Claude Enterprise, or access Claude through AWS Bedrock, Google Cloud Agent Platform, or Microsoft Foundry with ZDR.

https://support.claude.com/en/articles/15425996-data-retenti...

Re: There is minimal downside to switching to open models

#229

Post by 0xDEAFBEAD »

hdgvhicv wrote: post #117
earthnail wrote: post #90

Earlier quoted context omitted.

From all the large governmental institutions, the EU is the one currently holding up traditional western values. That gives it street cred in this subject.

https://www.theguardian.com/us-news/ng-interactive/2026/feb/... The age old joke; A Russian and an American are drinking at a bar The Russian says "I'm impressed by american propaganda. It's so subtle but effective." The american responds "What are you talking about, we don't do propaganda."

I'm of the opinion that there is considerably more wailing about US government propaganda than actual US government propaganda. People who reference supposed US government propaganda rarely provide much in the way of concrete examples. Probably because there are legal restrictions on covert propaganda in the US:

https://www.law.cornell.edu/wex/covert_propaganda

To be clear, I'm happy to grant that:

* The Pentagon won't provide jets for your war movie if your war movie portrays the US military in a bad light

* The US engages in information operations in foreign countries, e.g. discouraging people in the Philippines from getting the Chinese COVID vaccine

* Voice of America and similar US-government sponsored outlets are, in fact, sponsored by the US government

But the notion that covert, English-language US government propaganda is ubiquitous and effective seems like a half-baked, un-falsifiable conspiracy theory with little supporting evidence.

The internet is full of false or misleading claims about the US which go un-refuted. There's just way too much low-hanging fruit going un-picked here to believe that the USG is running massive English-language covert propaganda ops.

A specific example of a false anti-American claim which is extremely widespread: Many Europeans believe that the US promised to protect Ukraine in the 1994 Budapest Memorandum. This is false. We only promised to go to the UN Security Council, which we did. You can verify for yourself with a quick trip to the UN website, the memorandum is not very long: https://treaties.un.org/doc/Publication/UNTS/Volume%203007/P...

If the American government possessed the propaganda wizardry that people ascribe to it, I expect the entire internet would be well-acquainted with the actual contents of this memorandum. Instead, you have randos like me trying to fight a tsunami of misinformation (likely Ukrainian-origin) related to this memorandum, using only a shovel.

Re: There is minimal downside to switching to open models

#230

Post by beng-nl »

dminik wrote: post #187
himata4113 wrote: post #130

Earlier quoted context omitted.

The part that gets me about anthropic red lines is "of Americans", okay so the rest of the civilized world is up for grabs then? It's okay to destabalize allies with sabotaged tests (in machine learning) and data exfiltration outside America? What gets me the most is that they claim that the model should follow the https://www.anthropic.com/constitution and they claim that it's embedded into the model. However, syste…

Yeah, it was funny seeing a bunch of people going like "Anthropic is fighting for privacy" meanwhile I'm like "Uhh, what about the other 8 billion people?" On second thought, it's not funny.

As a thought experiment - such shocks (govt pressure to use models for bad purposes and govt excluding access to non-Americans) coming early in the ‘ai revolution’ will wake up the rest of the world sooner that they have to get their act together to stay competitive without relying on USA. Just like with nato.
Post reply on HN