The glaring number to me is only 5% of VC funds vs 52% in the US. That's 10x more opportunity despite roughly comparable economies. As long as that is true, it seems like it will always be impossible to get an organic startup industry working in the EU. Any startup that is any good will almost certainly end up getting a round of investment from the US and most likely move their base of operations there.
I wonder if recent US actions will start to influence this as there now appears to be more risk in sending your money to the US or founding your company there as a foreign entity than there used to be.
Not just startups or funding. Google alone has more AI compute than China and the EU combined.
There's no shortage of capital in Europe. But nobody wants to take the risk. Meanwhile in the US, people are putting 10-100x the capital at risk. So you can say what you want about it looking scary to you to invest in the US, but the people with capital to invest clearly don't see it that way.
More often when you hear VCs give interviews, they are saying the opposite: that never-ending EU regulations introduce more business risk than anything the US president could possibly do.
I’m not all that convinced on the regulation part.
Ultimately it’s all about investing money to create real assets that generate cash flows. One can side step regulations to some extent whilst developing a product (nobody cares/notices until you are actually growing fast) and then deal with regulations later. Uber already showed this and the leading AI firms are following the same act - having ripped off a lot of content but nobody threw a fit until a legit asset came out of it.
The funding allocation is a reflection of different cultures really.
I’m from the UK but there’s no way there’s the same density, drive, and hunger to take risk, to the extent that you find in US. Also there’s a lot more synergy in the US vs a fragmented Europe.
Yet the UK did produce Deepmind. Which you can look at both ways since it got gobbled up by Google. But it at least came from the UK.
I assume the article excludes the UK, I feel like UK has much more of a startup and VC scene than Europe does and I wonder if that is part of the issue : if you do want to create a startup, the London is a better place than mainland Europe. So even startups that for whatever reason won't take US funding still land outside Europe.
There's this sentiment in Germany that if you can't make in industry, you work for the government or - even worse - become a politician. It seems like Mistral took that to the next level; they can't compete so they do lobbying instead.
Being European, I love the idea of European AI labs. But I wish there was more competition.
That being said, as a German for example, I can't think of an AI company successfully training a competitive foundation model here. The copyright mafia would take your investor's money before you could even finish the first training run (hyperbole.)
I like the sentiment. Keeps a lot of people out of politics so the few can rob everyone else blind. No no, you don’t want this job it’s for loser hacks only.
In that case, hopefully the copyright mafia will take the money from US and Chinese LLM companies and redistribute it to the people who did the actual work fueling the models, such as myself.
I did not spend 10 years writing (A)GPL code for all of it to be stripped of its license, remixed and sold for profit.
Of course in a truly just world, the LLM companies who took my code without permission would beg with offers of owning a share of them because if I didn't consent their models would have to be destroyed.
Since my work is apparently so valuable that they just have to have it, it should count towards my retirement age too.
I lead a European owned and operated Data/AI company, Hopsworks. We are the only competitor to Databricks/Snowflake/etc based-in and from Europe. We can still compete, as we have a deep research/industry background (ex-MySQL and KTH folks).
Crossing the chasm is harder from here. Even if you build a great product (we are best at real-time AI) - we had a paper at SIGMOD 2024 where we showed higher thoughput/lower latency by a factor of 4-40X Databricks, AWS Sagemaker, and GCP vertex - we lack the echo chamber. (Try the mental exercise where Databricks' peers acknowledge massive over-performance through a peer-reviewed paper and imagine how much noise it would generate). Still, we can replace our competitor at their largest European customer, Zalando, for real-time AI. But it's a much harder slog than it should be due to the 10X lower round sizes (due to 10X smaller VC fund sizes). European pension funds place way more money in US VC funds than in EU VC funds - that is self-defeating.
I like the "european technology" movement not because of any nationalist ideas, but because it stimulates technological innovation and creates a new dynamic.
It's important to note that these efforts aren't nationalistic - they're multilateral. In fact, European nationalists are consistently trying to sabotage European efforts.
On the bright side: people seem to be moving away from such nationalistic ideas. Here's to Orban being the first of many defeats for them in the near future.
The paper has plenty of things, but seems to me the gist of it is that they want to secure the public procurement cash flow in lieu of venture capital? I may be reading this wrong, but all the labels and normative proposals listed point to that.
I've tried Mistral a few times, at first it seemed promising (though lagging) but at some point it seems like they stopped focusing on AI and shifted their focus to being a mouthpiece for EU policy and pushing for regulation. I can't really take any of their announcements seriously anymore.
A couple of weeks ago they were calling for a European AI tax to pay creatives.
Recordable CDs involved individuals making copies. AI is run by a couple of dozen people who give full access to other people's work, metered by the syllable.
It was never legal for massive corporations to record other people's work on CDs and sell them; that's the opposite of copyright. The comparison is absurd.
I used to use Mistral OCR, but found it was better just to write a program that sent the documents to Claude Sonnet to OCR instead. Claude is far better quality, better formatting and fewer errors.
I'm also using Voxtral TTS to try to replace OpenAI. It "works", but I've had problems with volume levels being radically different between different audio chunks. It doesn't seem to "understand the full text" the way OpenAI's voice models do, which can be more expressive. Voxtral sometimes sounds robotic in the reading. And some Voxtral TTS output contains music in the background occasionally, which suggests their training corpus isn't that clean. Try generating a personalized news podcast, and the intro may occasionally sound like the music for BBC News underneath....
As for not focusing on AI, there's this interview in the Big Technology Podcast 2 months ago, where the Mistral CEO says their main focus is on helping companies fine-train models for internal use, over being a general model builder.
"I sent money to the god knows how many trillion parameters fully closed source machine built on billions of dollars and it worked better than the model that I can self host from the guys next door"
yeah, no shit ? All you're saying is that you're happily locking yourself in to models you have zero control over and that Anthropic can fuck you over at any time.
However, yes, Mistral is not in the business of providing you with a perfect, general purpose model. They fine tune from their base models for specific tasks.
Mistral OCR 3 isn't open weights and isn't available for download. It's only available via API, or to companies via paid consulting with Mistral.
"For organizations with stringent data privacy requirements, Mistral OCR offers a self-hosting option. This ensures that sensitive or classified information remains secure within your own infrastructure, providing compliance with regulatory and security standards. If you would like to explore self-deployment with us, please let us know."
I used their OCR against a few hundred page PDF that was printed text but missing the OCR. It cost me $5 and was useless, it did worse than tesseract. That's how all my experience with mistral is
I don't really care about the content, but European software is also when you switch to the tab the energy consumption of your MacBook quadruples due to some inane animations.
Reads like asking for a EU handout. It touches on some visible issues in the single market, but most of what I've seen is not warranted. Eg. minimum spending quotas for AI work/integration/research, using European models (basically today = use Mistral), or carving residency process exceptions for AI researchers.
58 Minute reading time. I read the first dozen pages or so and I'm not sure what the goal of this thing is, why they wrote it, who they wrote it for? Is it aimed at European governments? Or companies? Or people? Or something else?
> This playbook provides a clear, actionable framework to position Europe as that powerhouse, accelerating AI development and adoption, attracting and retaining top talent, simplifying regulation without sacrificing values, and mobilizing public and private investment to build homegrown AI infrastructure. Only with it, Europe can ensure AI is not only developed in Europe, but for Europe and on Europe’s terms.
playbook for what?
> This document is not a theoretical exercise. It is a practical playbook
Seems quite theoretical? A lot of random statistics, and all the sections start with abstract empty claims in 'not x, y' slop format "Artificial intelligence is not an abstract promise. It is a tool that fulfills its potential when embedded in the real economy."
I'd love an executive summary of this for anyone who has AI tokens to spend (I've got some other stuff to get done with what remains of my quota this week). I'm not saying this report is bad, I'm just saying it didn't do enough to convince me to read it, and it has some patterns that would make me guess it's bad.
They essentially want a bunch of stuff and most importantly funding from the EU and using the FOMO angle to get them to act. This of course is not on merit. They see that no other lab in Europe really exists and are trying to seize an open opportunity.
I hope one day soon EU politicians ask themselves why it might be that there is only one single domestic AI lab that is basically an also-ran at this point.
> The question is no longer whether Europe can compete, ...
But it, too, do not ask myself this question any more.
Since EU seems to have already lost completely.
Even Proton's new local AI service uses Ollama, which was developed in USA and is pretty outclassed. Does HN say europe can do more than hope to catch up in five to ten years, if the race is still on then?
I don't understand why European providers can't just host open-weight models developed by the Chinese, or distill Google/OpenAI/Anthropic models to produce their own models on the the cheap.
Nobody acts like you need to invent steel to have a steel mill.
what's the point? The Chinese, Google and OpenAI are burning money and we get to use the service. Having AI providers locally barely creates any jobs, it's an easily substitutable service and it has (contrary to the claims by the AI crazies) very few national security implications.
Steel is a great example because we don't pollute our rivers with steel mills any more either. As Milton Friedman said, if someone wants to give you steel and you give them green sheets of paper, be thankful, nothing's easier to make than paper.
What are you losing, the bragging rights among nerds on the internet? Right now Americans are paying the energy bills, Sam Altman is paying for the compute, they make no money off it, and they're even publishing the models! So if push comes to shove, we can deploy them. But until then how is that not a great deal
> The mechanism consists of a revenue-based levy applied to all commercial providers placing AI models on the market or putting them into service in Europe, reflecting their use of content publicly available online. This levy would apply equally to providers based abroad, creating a level playing field. The proceeds would flow into a central European fund dedicated to investing in new content creation, and supporting Europe's cultural sectors.
Presumably Mistral is putting forth the most pro-AI position possible for the region.
So it sounds like anyone doing what you described is at risk of a tax that will make their offerings uncompetitive.
I'm a fan of Mistral, but this seems to be 80% "make Europe more startup-friendly in general" rather than anything specific to AI.
Given how un-startup-driven adoption of new technologies usually happens in Europe, I don't see this playbook becoming a cornerstone of how AI adoption will pan out in Europe.
I feel like Europe needs to remove barriers and let people do things freely rather than stuff like "Empower AI students". Ambitious people will naturally find a way to get stuff done and you just have to allow it to happen and not get in the way. At least "EU AI talent visa" sounds like it can work by removing barriers to relocate.
Other than you seeing a thing you want to see, why? There’s a pretty well known story behind that. Also, this post was written by a mensch, so have some respect.
I'd say that situation is representative of a very apropos mentality from a very apropos entity... but this reaction to even mentioning that is the most telling signal on if there's actually willpower to make what they're advocating for happen on the ground.
META used to hire in Netherlands until it stopped and left. I wondered why and a few times I heard that it was because it was hard to be dynamic in the country with the stubborn labour laws. The anecdote confirms my own bias but there seems to be not much mention on encouraging risk taking allowing dynamic entrepreneurship in this playbook, which leads me to believe this is a non issue?
The glaring number to me is only 5% of VC funds vs 52% in the US. That's 10x more opportunity despite roughly comparable economies. As long as that is true, it seems like it will always be impossible to get an organic startup industry working in the EU. Any startup that is any good will almost certainly end up getting a round of investment from the US and most likely move their base of operations there.
I wonder if recent US actions will start to influence this as there now appears to be more risk in sending your money to the US or founding your company there as a foreign entity than there used to be.
Not just startups or funding. Google alone has more AI compute than China and the EU combined.
There's no shortage of capital in Europe. But nobody wants to take the risk. Meanwhile in the US, people are putting 10-100x the capital at risk. So you can say what you want about it looking scary to you to invest in the US, but the people with capital to invest clearly don't see it that way.
More often when you hear VCs give interviews, they are saying the opposite: that never-ending EU regulations introduce more business risk than anything the US president could possibly do.
I’m not all that convinced on the regulation part.
Ultimately it’s all about investing money to create real assets that generate cash flows. One can side step regulations to some extent whilst developing a product (nobody cares/notices until you are actually growing fast) and then deal with regulations later. Uber already showed this and the leading AI firms are following the same act - having ripped off a lot of content but nobody threw a fit until a legit asset came out of it.
The funding allocation is a reflection of different cultures really.
I’m from the UK but there’s no way there’s the same density, drive, and hunger to take risk, to the extent that you find in US. Also there’s a lot more synergy in the US vs a fragmented Europe.
Yet the UK did produce Deepmind. Which you can look at both ways since it got gobbled up by Google. But it at least came from the UK.
I assume the article excludes the UK, I feel like UK has much more of a startup and VC scene than Europe does and I wonder if that is part of the issue : if you do want to create a startup, the London is a better place than mainland Europe. So even startups that for whatever reason won't take US funding still land outside Europe.
Imagine all the great investments the US VCs are missing outside of AI.
There's this sentiment in Germany that if you can't make in industry, you work for the government or - even worse - become a politician. It seems like Mistral took that to the next level; they can't compete so they do lobbying instead.
Being European, I love the idea of European AI labs. But I wish there was more competition.
That being said, as a German for example, I can't think of an AI company successfully training a competitive foundation model here. The copyright mafia would take your investor's money before you could even finish the first training run (hyperbole.)
Nobody can compete with trillions of dollars of capital if it’s a race for capital.
I'm not sure about that. Post GFC the ECB printed plenty of cash as well.
https://fred.stlouisfed.org/series/ECBASSETSW
I like the sentiment. Keeps a lot of people out of politics so the few can rob everyone else blind. No no, you don’t want this job it’s for loser hacks only.
In that case, hopefully the copyright mafia will take the money from US and Chinese LLM companies and redistribute it to the people who did the actual work fueling the models, such as myself.
I did not spend 10 years writing (A)GPL code for all of it to be stripped of its license, remixed and sold for profit.
Of course in a truly just world, the LLM companies who took my code without permission would beg with offers of owning a share of them because if I didn't consent their models would have to be destroyed.
Since my work is apparently so valuable that they just have to have it, it should count towards my retirement age too.
I lead a European owned and operated Data/AI company, Hopsworks. We are the only competitor to Databricks/Snowflake/etc based-in and from Europe. We can still compete, as we have a deep research/industry background (ex-MySQL and KTH folks). Crossing the chasm is harder from here. Even if you build a great product (we are best at real-time AI) - we had a paper at SIGMOD 2024 where we showed higher thoughput/lower latency by a factor of 4-40X Databricks, AWS Sagemaker, and GCP vertex - we lack the echo chamber. (Try the mental exercise where Databricks' peers acknowledge massive over-performance through a peer-reviewed paper and imagine how much noise it would generate). Still, we can replace our competitor at their largest European customer, Zalando, for real-time AI. But it's a much harder slog than it should be due to the 10X lower round sizes (due to 10X smaller VC fund sizes). European pension funds place way more money in US VC funds than in EU VC funds - that is self-defeating.
I like the "european technology" movement not because of any nationalist ideas, but because it stimulates technological innovation and creates a new dynamic.
It's important to note that these efforts aren't nationalistic - they're multilateral. In fact, European nationalists are consistently trying to sabotage European efforts.
On the bright side: people seem to be moving away from such nationalistic ideas. Here's to Orban being the first of many defeats for them in the near future.
"European nationalists" as in "nationalists which happen to reside in Europe", not "Chauvinistic European federalists", which feels like a rare breed.
The paper has plenty of things, but seems to me the gist of it is that they want to secure the public procurement cash flow in lieu of venture capital? I may be reading this wrong, but all the labels and normative proposals listed point to that.
Cynical. Start by allowing remote positions in Mistral across Europe.
I've tried Mistral a few times, at first it seemed promising (though lagging) but at some point it seems like they stopped focusing on AI and shifted their focus to being a mouthpiece for EU policy and pushing for regulation. I can't really take any of their announcements seriously anymore.
A couple of weeks ago they were calling for a European AI tax to pay creatives.
Sounds like the tax on recordable CDs.
https://en.wikipedia.org/wiki/Private_copying_levy
Ridiculous, and it didn't take money from the right people or give it to the right people. I would expect the same from an AI tax.
Recordable CDs involved individuals making copies. AI is run by a couple of dozen people who give full access to other people's work, metered by the syllable.
It was never legal for massive corporations to record other people's work on CDs and sell them; that's the opposite of copyright. The comparison is absurd.
Why do you say they "stopped focusing on AI"? I see a pretty consistent release of pretty good products - particularly in speech and OCR.
I used to use Mistral OCR, but found it was better just to write a program that sent the documents to Claude Sonnet to OCR instead. Claude is far better quality, better formatting and fewer errors.
I'm also using Voxtral TTS to try to replace OpenAI. It "works", but I've had problems with volume levels being radically different between different audio chunks. It doesn't seem to "understand the full text" the way OpenAI's voice models do, which can be more expressive. Voxtral sometimes sounds robotic in the reading. And some Voxtral TTS output contains music in the background occasionally, which suggests their training corpus isn't that clean. Try generating a personalized news podcast, and the intro may occasionally sound like the music for BBC News underneath....
As for not focusing on AI, there's this interview in the Big Technology Podcast 2 months ago, where the Mistral CEO says their main focus is on helping companies fine-train models for internal use, over being a general model builder.
https://www.youtube.com/watch?v=xxUTdyEDpbU&t=1357s
"I sent money to the god knows how many trillion parameters fully closed source machine built on billions of dollars and it worked better than the model that I can self host from the guys next door"
yeah, no shit ? All you're saying is that you're happily locking yourself in to models you have zero control over and that Anthropic can fuck you over at any time.
However, yes, Mistral is not in the business of providing you with a perfect, general purpose model. They fine tune from their base models for specific tasks.
Mistral OCR 3 isn't open weights and isn't available for download. It's only available via API, or to companies via paid consulting with Mistral.
"For organizations with stringent data privacy requirements, Mistral OCR offers a self-hosting option. This ensures that sensitive or classified information remains secure within your own infrastructure, providing compliance with regulatory and security standards. If you would like to explore self-deployment with us, please let us know."
https://docs.mistral.ai/models/ocr-3-25-12 https://mistral.ai/news/mistral-ocr-3
I used their OCR against a few hundred page PDF that was printed text but missing the OCR. It cost me $5 and was useless, it did worse than tesseract. That's how all my experience with mistral is
They are making some very good specialised models, like Voxtral
“European AI tax to pay creatives”
Love that idea.
I mean we already pay a tax when buying phones or storage because it's assumed we'll use it for piracy so why not.
If they want to improve but find that European regulations are the main obstacle, it makes sense they focus on that.
Maybe I wasn't very clear. It seems like they are in support of those regulations. They seem like an EU mouthpiece
I don't really care about the content, but European software is also when you switch to the tab the energy consumption of your MacBook quadruples due to some inane animations.
Reads like asking for a EU handout. It touches on some visible issues in the single market, but most of what I've seen is not warranted. Eg. minimum spending quotas for AI work/integration/research, using European models (basically today = use Mistral), or carving residency process exceptions for AI researchers.
58 Minute reading time. I read the first dozen pages or so and I'm not sure what the goal of this thing is, why they wrote it, who they wrote it for? Is it aimed at European governments? Or companies? Or people? Or something else?
> This playbook provides a clear, actionable framework to position Europe as that powerhouse, accelerating AI development and adoption, attracting and retaining top talent, simplifying regulation without sacrificing values, and mobilizing public and private investment to build homegrown AI infrastructure. Only with it, Europe can ensure AI is not only developed in Europe, but for Europe and on Europe’s terms.
playbook for what?
> This document is not a theoretical exercise. It is a practical playbook
Seems quite theoretical? A lot of random statistics, and all the sections start with abstract empty claims in 'not x, y' slop format "Artificial intelligence is not an abstract promise. It is a tool that fulfills its potential when embedded in the real economy."
I'd love an executive summary of this for anyone who has AI tokens to spend (I've got some other stuff to get done with what remains of my quota this week). I'm not saying this report is bad, I'm just saying it didn't do enough to convince me to read it, and it has some patterns that would make me guess it's bad.
Indeed very slop-feeling "whitepaper", might as well be written by chatgpt/claude because it has the tropes.
Multiple sections have expandable subsections for more details on proposals.
It’s a load of nonsense.
They essentially want a bunch of stuff and most importantly funding from the EU and using the FOMO angle to get them to act. This of course is not on merit. They see that no other lab in Europe really exists and are trying to seize an open opportunity.
I hope one day soon EU politicians ask themselves why it might be that there is only one single domestic AI lab that is basically an also-ran at this point.
The site starts with
> The question is no longer whether Europe can compete, ...
But it, too, do not ask myself this question any more. Since EU seems to have already lost completely.
Even Proton's new local AI service uses Ollama, which was developed in USA and is pretty outclassed. Does HN say europe can do more than hope to catch up in five to ten years, if the race is still on then?
I don't understand why European providers can't just host open-weight models developed by the Chinese, or distill Google/OpenAI/Anthropic models to produce their own models on the the cheap.
Nobody acts like you need to invent steel to have a steel mill.
what's the point? The Chinese, Google and OpenAI are burning money and we get to use the service. Having AI providers locally barely creates any jobs, it's an easily substitutable service and it has (contrary to the claims by the AI crazies) very few national security implications.
Steel is a great example because we don't pollute our rivers with steel mills any more either. As Milton Friedman said, if someone wants to give you steel and you give them green sheets of paper, be thankful, nothing's easier to make than paper.
What are you losing, the bragging rights among nerds on the internet? Right now Americans are paying the energy bills, Sam Altman is paying for the compute, they make no money off it, and they're even publishing the models! So if push comes to shove, we can deploy them. But until then how is that not a great deal
I don’t know either, but playbook item #20 has:
> The mechanism consists of a revenue-based levy applied to all commercial providers placing AI models on the market or putting them into service in Europe, reflecting their use of content publicly available online. This levy would apply equally to providers based abroad, creating a level playing field. The proceeds would flow into a central European fund dedicated to investing in new content creation, and supporting Europe's cultural sectors.
Presumably Mistral is putting forth the most pro-AI position possible for the region.
So it sounds like anyone doing what you described is at risk of a tax that will make their offerings uncompetitive.
So why even bother?
I'm a fan of Mistral, but this seems to be 80% "make Europe more startup-friendly in general" rather than anything specific to AI.
Given how un-startup-driven adoption of new technologies usually happens in Europe, I don't see this playbook becoming a cornerstone of how AI adoption will pan out in Europe.
Mistral is the only AI bot I haven’t blocked from my server.
How?
Do they respect robots.txt?
I feel like Europe needs to remove barriers and let people do things freely rather than stuff like "Empower AI students". Ambitious people will naturally find a way to get stuff done and you just have to allow it to happen and not get in the way. At least "EU AI talent visa" sounds like it can work by removing barriers to relocate.
Interesting reading this with
> Tell HN: docker pull fails in spain due to football cloudflare block
next to it on the front page
Other than you seeing a thing you want to see, why? There’s a pretty well known story behind that. Also, this post was written by a mensch, so have some respect.
I'd say that situation is representative of a very apropos mentality from a very apropos entity... but this reaction to even mentioning that is the most telling signal on if there's actually willpower to make what they're advocating for happen on the ground.
META used to hire in Netherlands until it stopped and left. I wondered why and a few times I heard that it was because it was hard to be dynamic in the country with the stubborn labour laws. The anecdote confirms my own bias but there seems to be not much mention on encouraging risk taking allowing dynamic entrepreneurship in this playbook, which leads me to believe this is a non issue?
But messagebird is another example.
I find it so funny that the people leading AI today have names that could be the characters of a bad scifi novel.
Amodei (love of god) Altman (alternative to man?) Arthur Mensch fighting from the ethical side
this whole timeline feels like a Bad Cyberpunk novel, Trump, Musk etc