I think people from the US often aren't aware how many companies from the EU simply won't risk losing their data to the providers you have in mind, OpenAI, Anthropic and Google. They simply are no option at all.
The company I work for for example, a mid-sized tech business, currently investigates their local hosting options for LLMs. So Mistral certainly will be an option, among the Qwen familiy and Deepseek.
Mistral is positioning themselves for that market, not the one you have in mind. Comparing their models with Claude etc. would mean associating themselves with the data leeches, which they probably try to avoid.
We're seeing the same thing for many companies, even in the US. Exposing your entire codebase to an unreliable third party is not exactly SOC / ISO compliant. This is one of the core things that motivated us to develop cortex.build so we could put the model on the developer's machine and completely isolate the code without complicated model deployments and maintenance.
Mistral are mostly focusing on b2b, and for customers that want to self-host (banks and stuff). So their founders being from Meta, or where their cloud platform are hosted, are entirely irrelevant to the story.
The fact they would not exist without the leeches and built their business on the leeches is irrelevant.
Pan-nationalism is a hell of a drug: a company that does not know you exist puts out an objectively awful release, and people take frank discussion of it as a personal slight.
Those who crawled the web without consent, and then put their LLM in a blackbox without attribution, with secret prompt and secret weights -- ie. all of this without giving back, while creating tons of Co2. Those are the leeches.
Ah, so "crawled the web without consent, and then put their LLM in a blackbox without attribution" is not being a leech once you release the weights of an underperforming model using someone else's arch.
At the very least it is a step in the right direction. Can't say the same for these proprietary models. And guess which country has all these proprietary models? USA.
I didn't mean to imply US bad EU good. As such, this isn't about which passport the VCs have, but about local hosting and open weight models. A closed model from a US company always comes with the risk of data exfiltration either for training or thanks to CLOUD Act etc (i.e. industrial espionage).
And personally I don't care at all about the performance delta - we are talking about a difference of 6 to at most 12 months here, between closed source SOTA and open weight models.
It's wayyyy to early in the game to say who is out-executing whom.
I mean why do you think those guys left Meta? It reminds me of a time ten years ago I was sitting on a flight with a guy who works for the natural gas industry. I was (cough still am) a pretty naive environmentalist, so I asked him what he thought of solar, wind, etc. and why should we be investing in natural gas when there are all these other options. His response was simple. Natural gas can serve as a bridge from hydrocarbons to true green energy sources. Leverage that dense energy to springboard the other sources in the mix and you build a path forward to carbon free energy.
I see Mistral's use of US VCs the same way. Those VCs are hedging their bets and maybe hoping to make a few bucks. A few of them are probably involved because they're buddies with the former Meta guys "back in the day." If Mistral executes on their plan of being a transparent b2b option with solid data protections then they used those VCs the way they deserve to be used and the VCs make a few bucks. If Europe ever catches up to the US in terms of data centers, would Mistral move off of Azure? I'd bet $5 that they would.
> Russia will conquer Ukraine, any other prediction at this point is absurd.
Are you sure? They are advancing, sure, put look what they paid for to achieve this: 300k dead, 700k wounded, depletion of their souvereign wealth fund, 20%+ inflation, lower oil production and so on.
Unfortunately, yes. USA is doing everything but openly support Russia at this point too. It could have been different if Ukraine got proper support, but instead it is being undermined.
Europe could do more, but at least most states dont play for Russia (Hungary and Slovakia excepted).
I think we may be at peak Trump though which will limit his power to bail out Putin. The midterms won't go well, the Epstein stuff is embarrassing, the Republicans are starting to get unruly.
> They are advancing, sure, put look what they paid for to achieve this: 300k dead, 700k wounded, depletion of their souvereign wealth fund, 20%+ inflation, lower oil production and so on.
Russia is a totalitarian dictatorship led by the communist Putin. As if communist dictators care. Look at North Korea, it's just the results of an unremarkable year.
I think literally nobody knows the price either side is paying right now. And I do mean literally, including Trump, Putin, and Zelensky. The fog of war applies to participants, let alone outsiders who are basing our views on figures and claims that obviously going to be driven heavily by propaganda.
But beyond this, I don't think this war is about Ukraine anymore than a war in Taiwan will be about Taiwan. It's little more than a proxy for hegemony in both cases. Russia did not want NATO parked in their Achille's heel of the Ukrainian flatlands. NATO did, and we pushed forward against endless threats of it being a redline, essentially as a means of indirectly imposing our will on Russia and establishing a hierarchy of dominance.
And similarly, for those that don't the Taiwan-China history - the Mao led Chinese revolution was a success. The existing government of mainland China fled to Taiwan where they brutally oppressed the locals, in an era known as the 'white terror' [1], and established power through 40 years of martial law. And of course we backed them, solely to use them as a weapon against China, because geopolitics.
This is why these wars are so important for the participants. The US couldn't care less about Ukraine, but withdrawing without ruining our ability to militarily threaten other peer or near peer countries is difficult. And similarly the last thing Russia needs is more land, but if they never act on claims of red lines, then they can never expect their interests to be considered in the case of a conflict in interests between them and the West.
I don't agree on the Russia Ukraine motivations. Ukraine is not part of NATO and was not going to become part of NATO. There were already two NATO countries bordering Russia near Moscow and St P if NATO had wanted to invade which they had no thoughts of doing. Russia lies constantly on this stuff. I think they basically regarded Ukraine their land as part of the Russian empire they were restoring.
It's not about immediate intentions, but about strategic options. Imagine Russia decided to form a military alliance with Mexico with the expected intention of deploying weapons on the Mexican border. If Mexico agreed to this, it would take approximately 0 seconds before the US invaded them under some whimsical pretext (drug gangs probably) and overthrew their government to prevent this. In fact this is, more or less, what the Cuban Missile Crisis was where we were willing to bring the world to the brink of nuclear annihilation over it, and that was an even lighter weight version of this event since there isn't even a land route from Cuba to the US obviously!
But in this scenario would you think Russia deploying weapons in Mexico is a precursor to them invading? Or that the US would be worried about that? Obviously not. Neither was Cuba. But it gives an adversarial power a tremendous strategic edge, while you get less than nothing out of it since it reduces your 'power' in the relative strategic balance of countries.
> Imagine Russia decided to form a military alliance with Mexico with the expected intention of deploying weapons on the Mexican border.
It would be a very foolish idea, because it's no longer the Napoleonic era. Concentrating your forces close to adversary's border makes them easy targets for destruction by long-range artillery and airstrikes. The Finnish chief of defence forces recently made the same remark when the Russians moved their weapons closer to Finland for intimidation: "It only makes them easier for us to destroy."
> In fact this is, more or less, what the Cuban Missile Crisis was
Not at all. The Cuban missile crisis was only about nuclear missiles. The USSR continued to provide a large number of conventional weapons to Cuba, including submarines and fighter jets, until it collapsed in 1991, without any of your invasion fantasies coming true.
It is a Soviet-built MIG-23 fighter jet carrying Cuban insignia. MIG-23 first flew 5 years after the missile crisis and the first batch was delivered to Cuba in 1978.
> It's not about immediate intentions, but about strategic options. Imagine Russia decided to form a military alliance with Mexico with the expected intention of deploying weapons on the Mexican border.
The problem with pretending this analogy is relevant as a justification (or at least an "other people would have one the same thing" argument, which isn't really a justification to start with) of the Russian invasion of Ukraine (besides the fact that it relies on dubious assumptions about a counterfactual) is that the only reason Ukraine resumed its long-abandoned pursuit of relations with NATO was a direct result of the invasion by Russia in 2014.
Ukraine had been striving repeatedly to join NATO until 2010. That's when Yanukovych, who generally leaned more East than West, took power. Ukraine dropped its NATO ambitions under his leadership and re-affirmed themselves as a neutral state. Then he was overthrown, in an action directly backed by the US with John McCain, Victoria Nuland, and others literally on the ground in Ukraine giving speeches and riling up protesters come rioters, almost certainly with further black ops organizing going on behind the scenes.
Following Yanukovych's successful overthrow figures favorable to the US/UN/EU, including those hand picked by Victoria Nuland in her leaked conversation, ended up in power. In fact the person Nuland hand picked for Prime Minister, Arseniy Yatsenyuk, was one of the authors of Ukraine's initial formal request for a membership action plan from NATO.
Can you tell me that you genuinely think that if Russia hadn't annexed Crimea (which happened after all of the above) that Ukraine would have chosen to stay "neutral" in this context? And I put neutral in quotes because what does that even mean when one bloc is driving the successful overthrow of democratically elected leaders and hand picking new ones? Imagine Lavrov et al were on the ground encouraging pro Russian protesters to topple the Ukrainian government (alongside comparably likely black ops organizing behind the scenes), they ended up successful, and then leaders hand-picked by him end up in power. Is that somehow still just Ukraine deciding their own fate?
There's a huge problem with this narrative. The Russian government's public tender database shows that they ordered the production of campaign medals for the invasion of Crimea months before any of this happened. Oops.
I still think Ukraine wasn't primarily about Russia's military security though. I mean the US/Nato could stick missiles in Estonia if they wanted.
It may have been about political security. If Ukraine which is basically at least part Russian had become a prosperous democracy on Russia's doorstep it would make it harder for Putin to justify his autocracy. In fact that one may come to pass.
It's not about missiles in this case. That's a strategic battle that Russia has largely already lost, though the advent of highly capable ICBMs/MIRV/etc with hypersonic maneuvering also makes vicinity less relevant in modern times. In this case it's about a land route for invasion and subsequent logistics. There are already NATO countries bordering Russia, but the land between them is extremely unfavorable - swamps, forests, and so on. It's simply not fit for what would be a large scale conflict.
Invasion into Russia would ideally go through Belarus, which is part of the reason that Belarus is such a critical ally for Russia, and now even hosts their nuclear weapons. Since that's not possible, the second best route (and third and forth and...) is through Ukraine, likely towards Kursk or Belgorod.
There's even something of an equal but opposite here on NATO's side - the Suwalki corridor [1]. It's a narrow stretch of land between Belarus and Kaliningrad (a Russian exclave) that, if controlled, would cut off the Baltic states from NATO. So if war ever breaks out between NATO and Russia, it would be a key strategic point and unsurprisingly, it's been heavily fortified by NATO - there are even hundreds of American troops there.
The idea of an invasion of Russia from Europe is utter nonsense and completely detached from reality. Tell it to Russian military experts and you will get sighs and eyerolls in response. Not even Russian military exercises like Zapad simulate such a scenario. On the ground, the border remains completely open - you can walk straight into Russia (and lost mushroomers often do so by accident) because there isn't even a chainlink fence or a cleared sand strip marking the border.
Contrast that with the European countries that actually fear an invasion: they are preparing bridges for demolition, scouting suitable areas for minefields, digging anti-tank ditches, installing reinforced pillboxes and bunkers. Last week, Latvian media reported that the government is even considering tearing up railways near the Russian border to slow the invading force.
The scenarios the Russians are preparing for include, for example, mass unrest in Belarus that would lead to Russia invading the country to keep its dictator in place, like they did in Czechoslovakia in 1968 and in Hungary in 1956. In 2020, this almost happened in Belarus over fraudulent elections and mass protests that were ultimately suppressed without requiring a "brotherly military intervention" by Russia.
> it would take approximately 0 seconds before the US invaded
Very unlikely.
Also Mexico wasn’t never exactly that aligned diplomatically and politically with the US to begin with.
Russia on the other hand views that it has some inherent right to subjugate and dominate all of their neighbors and turn them into puppet states if not outright annex them.
> In fact this is, more or less, what the Cuban Missile Crisis
In fact this is outright drivel. The US hardly viewed Russia as their actual opponent before 2014-22. Remember Romney- Obama debate (and Obama generally bending over backwards to appease Putin most of the time).
> Russia did not want NATO parked in their Achille's heel of the Ukrainian flatlands
Russia (i.e. Putin but also Russians in general) wanted to rebuild their empire from the beginning. Anything else is just an excuse.
> interests between them and the West
Of course this conflict has been mostly one sides till the 2014, with Obama and Merkel bending over backwards to appease Putin.
Also the implication that Russia has some God given right over dominion of half of Eastern Europe is a bit appealing..
> our will on Russia and establishing a hierarchy of dominance.
That is a very Ruso-Imperialist mindset. A society pretty permanently stuck in the 1800s politically and psychologically… e.g. Germany, France, Britain were somehow able to step over their ambitions and are doing relatively fine (even without having millions of foreigners subjugate)
Thank you for repeating Russian propaganda. But the truth is that Ukraine is sovereign nation and has every right to decide their future and give a fuck about Russia feelings. Russia is the aggressor and blaming anything on NATO is laughable propaganda.
> In all honesty, would you hold that argument if Mexico decides to host Russian or Chinese troops?
Ukraine wasn't hosting foreign troops (except Russian troops, some of whom were were the spearhead of the invasion) when the Russo-Ukrainian war started with the Russian invasion in 2014.
(They did start hosting some that were involved in training and advisory assignments after the war started and before the major escalation in 2022, but those can hardly justify the war which started with the 2014 invasion.)
In all “honest” how is that relevant when Ukraine never did that nor was US willing to deploy their troops there to begin with. To what end? Not a single US administration between 1990 and 2022 was particularly antagonistic or expansionist towards Russia..
> The USA is still THE powerhouse economy of the world.
Things look decidedly different if you exclude the ad companies (Google, Meta, ...) and associated shovel sellers, see the WaPo article about the S&P 493 from a few days ago.
Not to discount physical infrastructure, but the world is quite digital these days and being at the absolute top of the software + associated techs economy is nothing to sniff at.
OTOH, why don't they ship good enough products? To me all of OpenAIs recent investments strongly suggest they hit a dead end with their current LLM approach. After all, if they knew the path ahead for GPT looks great, why don't they invest into training the next big thing instead of doing datacenters with the intention of renting them out?
I have two sets (one is a 4K, and the other, a regular set). It happens on both. The crashes are different, though, for each app. Usually, it’s the app falling into a fugue state, at some point, as I am channel-surfing (I do that, a lot).
I should also qualify that it’s not really “every single one” (that’s hyperbole). It’s the ones that I routinely use (Apple, Amazon, Hulu, and Netflix).
I’ll probably get another one, sooner or later, but I’ve been waiting for them to release a new version[0],
Interestingly enough the same happens on both Spotify's desktop and Chromecast / Apple TV application when you "song surf".
If you rapidly keep skipping through a song, then to the next one and repeat, the performance will keep on tanking. After a while it'll take 5-10 seconds just to load a song, going to other UI sections will take 3-5 seconds to load, and eventually the application completely locks up and soft reboots itself.
Javascript garbage collector wouldn't take so long... probably it's launching a new thread or process (for example, DRM decryption) for every song, and there becomes too many of them. Or it runs out of memory and starts swapping.
To drive this point further: The one stable causal relastionship relevant here is the one between inequality and crime. Reduce inequality in ways
0xbadcafebee suggested would reduce inequality - though probably not in sizes measureable after a few years.
It's probably reasonable to take a step back here and ask: Why is this not a universal problem? It's not as if every juristication outside the US simply lets criminals run away.
Do you have an example of a good implementation of ai captions? I've only experienced those on youtube, and they are really bad. The automatic dubbing is even worse, but still.
On second thought this probably depends on the caption language.
I'm not going to defend the youtube captions as good, but even still, I find them incredibly helpful. My hearing is fine, but my processing is rubbish, and having a visual aid to help contextualize the sound is a big help, even when they're a bit wrong.
Your point about the caption language is probably right though. It's worse with jargon or proper names, and worse with non-American English speakers. If we they don't even get right all the common accents of English, I have little hope for other languages.
Automatic translation famously fails catastrophically with Japanese, because it's a language that heavily depends on implied rather than explicit context.
The minimal grammatically correct sentence is simply a verb, and it's an exercise to the reader to know what the subject and object are expected to be. (Essentially, the more formal/polite you get, the more things are added. You could say "kore wa atsu desu" to mean "this is hot." But you could also just say "atsu," which could also be interpreted as a question instead of a statement.)
Chinese seems to have similar issues, but I know less about how it's structured.
Anyway, it's really nice when Japanese music on YouTube includes a human-provided translation as captions. Automated ones are useless, when it doesn't give up entirely.
I assume people talk about transcription, not translation. Translation in youtube ime is indeed horrible in all languages I have tried, but transcription in english is good enough to be useful. However, the more technical jargon a video uses, the worse transcription is (translation is totally useless in anything technical there).
Automatic transcription in English heavily depend on accent, sound quality, and how well the speaker is articulating. It will often mistake words that sound alike to make non-sensible sentences, randomly skip words, or just inserts random words for no clear reason.
It does seem to do a few clever things. For lyrics it seem to first look for existing transcribed lyrics before making their own guesses (Timing however can be quite bad when it does this). Outside of that, AI transcribed videos is like an alien who has read a book on a dead language and is transcribing based on what the book say that the word should sound like phonetically. At times that can be good enough.
(A note on sound quality. It not the perceived quality. Many low res videos has perfectly acceptable, if somewhat lossy sound quality, but the transcriber goes insane. It likes prefer 1080p videos with what I assume much higher bit-rate for the sound.)
In the times I have noticed the transcription be bad, my speech comprehension itself is even worse. So I still find it useful. It is not substitution for human created (or at least curated) subtitles by any means, but better than nothing.
Do you have an example? YT captions being useless is a common trope I keep seeing on reddit that is not reflected in my experience at all. Feels like another "omg so bad" hyperbole that people just dogpile on, but would love to be proven wrong.
There are projects that will run Whisper or another transcription service locally on your computer, which has great quality. For whatever reason, Google chooses not to use their highest quality transcription models on YouTube, maybe due to cost.
I use Whisper running locally for automated transcription of many hours of audio on a daily basis.
For the most part, Whisper does much better than stuff I've tried in the past like Vosk. That said, it makes a somewhat annoying error that I never really experienced with others.
When the audio is low quality for a moment, it might misinterpret a word. That's fine, any speech recognition system will do that. The problem with Whisper is that the misinterpreted word can affect the next word, or several words. It's trying to align the next bits of audio syntactically with the mistaken word.
Older systems, you'd get a nonsense word where the noise was but the rest of the transcription would be unaffected. With Whisper, you may get a series of words that completely diverges from the audio. I can look at the start of the divergence and recognize the phonetic similarity that created the initial error. The following words may not be phonetically close to the audio at all.
Ah yes, one of the standard replies whenever anyone mentions a way that an AI thing fails: "You're still using [X]? Well of course, that's not state of the art, you should be using [Y]."
You don't actually state whether you believe Parakeet is susceptible to the same class of mistakes...
It's an extremely common goalpost-moving pattern on HN, and it adds little to the conversation without actually addressing how or whether the outcome would be better.
Try it, or don't. Due to the nature of generative AI, what might be an issue for me might not be an issue for you, especially if we have differing use cases, so no one can give you the answer you seek except for yourself.
Apparently it's also shit. There was a discussion about it a few days ago that contains multiple project maintainers pointing out deepwiki didn't get their repos at all https://news.ycombinator.com/item?id=45884169
The company I work for for example, a mid-sized tech business, currently investigates their local hosting options for LLMs. So Mistral certainly will be an option, among the Qwen familiy and Deepseek.
Mistral is positioning themselves for that market, not the one you have in mind. Comparing their models with Claude etc. would mean associating themselves with the data leeches, which they probably try to avoid.
reply