Posted by NotInOurNames 10/28/2025
I used the 9B Instruct version, from the small models, it was the one with the best Latvian knowledge out there, bar none. GPT-OSS 20B and Qwen3 30B A3B and similar ones weren't even close.
That said, the model itself was a little bit dumb and not something you'd really use for programming/autocomplete or tool calling or anything like that, which also presented some problems - even for processing text, if you need RAG or tool server calls, you need to use something like Qwen3 for the actual logic and then pass the contents to EuroLLM for translation/formatting with the instructions, at which point your n8n workflow looks a bit messy and also you have to run those two models instead of only one.
Meanwhile, the best cloud model for Latvian that I've found so far was Google Gemini 2.5 Pro, but obviously can't use cloud models in certain on-prem use cases.
I have to specifically tell something like this: “do you known Lithuanian language”, then it starts replying in Lithuanian
While he is great at converting his influencer status to income in his micro-SaaS projects, I don't think running ad-fueled browser games on state-sponsored super computer should be really aim of these grant programs.
I’ve seen this in practice with innovation subsidies: small startup has to spend a lot of time to go though all the hoops, taking up significant resources, to get modest compensation for actual innovative work. Larger business hires an external agency to do everything for them, to make silly proposals “convert website from jquery to react” as innovation, and gets thousands upon thousands of FTE-hours in subsidy compensation hours for it.
The EU is also great at creating a heavy regulatory environment. Which entrenches existing incumbents. So the EU creates barriers that favor big companies, then tries to fix it with grants that... also favor big companies.
And then everyone's surprised that there's no innovation in Europe.
From all the world's companies worth over 100B$ there's only one European company - SAP, founded 50 years ago. [1]
[1] https://www.economist.com/briefing/2021/06/05/once-a-corpora...
Would you rather have an economy of SAPs or an economy of Teslas?
In part probably because it's harder to become a monopoly in the EU.
This x2. A close friend of mine works at a major EU HW tech company and his job is wearing a suit, going to dinner parties and rubbing shoulders with high level local and national bureaucrats to convince them to fund X, Y, Z projects, none of which result in any major commercial success or ROI for the governments because the money they get is not enough to make new successful products, but hey, he's loving his job which gives him amazing job security against the waves of layoffs the company went through due to falling of sales, plus the networking he gets out of that is invaluable.
So at least some people are enjoying the gravy train while it lasts. But that's why a lot of EU tech companies immediately go to the US first before opening up to the EU market. US VSs are more generous with their cheque books than EU governments and investors, plus the 300 million people single consumer market speaking English as the common language and all that.
All these while the EU is running out of funds and in a process of de-industrialization. There should be an independent corruption investigation on Brussels.
It's bureaucracy, often bordering with stupidity. You may need advisors to navigate all their forms & processes. But it certainly isn't "pals-only" type of deal.
On the other hand - is it harder than getting VC funding? For seasoned founder with reputation - probably. For fresh startup - probably not.
The probability of getting a Horizon Europe grant allegedly (not official stats) is about 8.5% according to some friends, which may seem low. You need to write 70 pages following a Word template and the key goal is to cover answers to a large number of questions. Each proposal gets various grades across a range of dimensions, which get added up and if you obtain at least 13 out of a possible 15 points, you are eligible to get funded, read: "You will get funded if there is enough money." Often, there are several proposals that justly achieve 15/15, and because of that, many prosals that have 14 points and all proposals that have less may not get funded, simply because there just is not enough total funding available to fund all the technically eligible proposals. Having judged many proposals in AI / ML / search / "big data" / language technology etc. I recommend optimizing recall, i.e. aspiring completeness.
The application process is not easy, but you can get help: there are support agency in each member country, free online Webinars to help, hotline help desks as well as an ecosystem of paid consultants that typically charge about 3k€ to vet a proposal for you if you need that kind of service (I never used it).
The process is neutral and conducted professionally and with external oversight (consultants are hired as "rapporteurs" that report on process/procedural integrity in additional to the actual reviewers). I value the research officers of the EC as people of high competency, integrity and motivation (research money is tax payers money so it should be spent carefully).
In comparison, VC (and even more so business angel) funding is achievable with much less formal apparatus, often a short business plan and a convincing slide deck and demo can get people to a partner meeting if the time is right. But the criteria and process are much different, and ideas ready for public research grants are typically too early for VCs (but the EC wants to foster the creation VC-funded startups resulting from the disseminated research).
> The EIC welcomes applications from innovators in all EU Member States and countries associated to the Horizon Europe programme. It particularly welcomes applications from startups and SMEs with female CEOs.
So that's one that strongly pushes that angle. But the one I saw I clearly remember saying "female co-founder", not CEO.
highly doubt, the whole thing about the success of the US west coast is that they are&were willing to fund unproven upstarts.
The point being that, as soon as public dollars are on the table, people expect perfection. Anything less is waste, fraud, and abuse.
There's literally no winning. Want to make sure the money is allocated right? Bureaucracy. Want to not do that? Waste, fraud, and abuse.
Except that Apple, Intel, Tesla, etc have all received US government investment [1]. TSMC is a product of the Taiwanese state! Government investment can be done well, and seeds excellent companies.
[1]: https://www.sba.gov/blog/2024/2024-02/white-house-sba-announ...
The comment you're replying to is tainted with the survivorship bias. We see successful companies that got government funding, but not the opposite. Maybe we'd have more innovation and competition without government picking these specific winners.
Ironically, one of the companies you mentioned (Apple) now operates in an environment with very little competition and regularly faces antitrust claims.
Government picking winners may actually reduce competition in the long run. The key difference: when private money picks wrong, it's their loss. When government picks wrong, it's taxpayer money.
https://www.politico.eu/article/ombudsman-slams-commission-f...
Yes, some of the questions are weird, but I'd really rather write a bit confirming that the AI system being developed isn't going to be racist or Skynet than jump through some other hoops that exist (and that absolutely includes VC due diligence). The actual biggest issue with European funds is they get way more competent applications than they can fund anyway.
When you talk to most EU business owners, even in tech, the limiting factor isn't regulations. This being the #1 reason is such a tired trope.
Ironically, China has in some ways a bigger regulatory burden when it comes to software, as there if the government doesn't approve the business is dead in the water. I doubt that Klarna would've gotten off the ground there, for one, I could see them being shut down much earlier there. In the EU only now very slowly are some governments even starting to talk about some weak measures around their business model. But I've never, not once in my life, heard "Chinese software companies can't get off the ground due to the regulatory burden".
The same people who clamor about the EU regulations are the ones who hate on the EU for their protectionist measures against US tech. Yet another bout of irony here - China's software industry has flourished exactly thanks to 10 times stronger protectionist measures against US tech. So has Korea's, and their protectionism has never even been anywhere on the China level, more inbetween EU and China. No, if there's anything that would help, it's much more tech protectionism in the EU.
Pieter Levels is at the end of the day an influencer, not a serious founder.
I have a tech startup in Estonia and I agree. To me the biggest limiting factor is lack of funding.
In the US, you can get VC across state borders and VC can invest in any US company regardless of their location. Not in the EU.
Investing in an Estonian startup from, say, the Netherlands is near impossible. And even if managed, now that NL investor has to repeat the entire process for Poland, Spain, Malta and other countries.
EU Made Simple explains this very well https://www.youtube.com/watch?v=1RqYws1JAuI
If anything this would be due to Dutch regulations rather than Estonian ones. So that would be more of a problem for setting up VCs in the EU rather than attracting capital in general, as the same would apply to the rest of the world. Is it easier for a US or Singapore VC to invest in an Japanese startup vs an Estonian startup? As far as I know the answer is a "no", but I'm happy to be proven wrong.
A texas VC can invest in a startup in Chicago and a startup in Florida can raise money from NY and so on.
> If anything this would be due to Dutch regulations rather than Estonian ones
Yes, because there's no unified capital market. You describe an effect not the cause. With 27 member states, if such regulations are bi-directional, you'd need 351 of such "regulations". If one way, you'd need 702.
In the US with 51 states, there aren't a total of 1,275 "regulation contracts" between all states, there's one, and it's federal.
> Is it easier for a US or Singapore VC to invest in an Japanese startup vs an Estonian startup?
This was my main question, and it looks like the answer is indeed no.
Still, the existence of the EU is (to bring peace by) lowering, mostly economic, barriers that traditionally exist between countries.
And the comparison in this threat was "between EU and US" (in europe Thing is hard... in Europe you cannot... etc). When people use "The EU" or "Europe" in this sense, they consider it as a "Single thing". Europe, or the EU, has a great disadvantage for startups over the US. You say that is because Europe is not a "thing" like a country is. I say, that is because Europe seen as a "thing" lacks some critical infrastructure that "the thing called US" has - or even "the thing Brazil" or "China" do have, internally.
And you register online.
Opening a company in Estonia is very cheap but in Spain the manager/CEO needs to be an "autónomo" (like a self-employed tax status). This costs thousands of Euros per year. Something like 2,400-30,000 Euros per year, every year, forever.
The article is unclear, but is probably referring to making it easier for startups to offer products in other EU countries.
It's in very early stages, so info is very scattered. More info, for example, here: https://www.loyensloeff.com/insights/news--events/news/the-2...
And that's just one of the problems (many of the problems have nothing to do with European bureaucracy)
Secondly, what's easier besides VC funding? If it's VC funding, the disparity there has nothing to do with regulations - guess how much VC funding the non-EU rest of the world gets.
It’s a distant memory to me now, I’m building a company and so much has happened that the details of this decision have faded away. But, between the AI act and GDPR, there’s a set of potential traps laid out for you to step into, along with reams of paperwork. All that requires lawyers and compliance consultants to help you figure it out, and that’s way too much for a fledgling startup.
I think it said it all that the AI regulations were written before there was really anyone to regulate. Why would I want to pour my heart and soul into a system that’s geared to find ways to stop me from building?
Anyway, it’s no longer relevant to me: I’m gone and I don’t have to worry about it anymore.
I don't want to exchange my freedoms for your shareholder value, thank you.
Yet startups here have managed to compete with US bigtech incomparably better than the EU. Shows that tough privacy laws have nothing to do with it.
> All that requires lawyers and compliance consultants to help you figure it out, and that’s way too much for a fledgling startup.
It also really just doesn't unless you're doing really shady stuff, in which case, good. The huge majority of startups don't need a lawyer to deal with GDPR.
Which part is easier? That you have 50 different states with slightly varying laws to consider (e.g. Californian Data protection)? That you have a byzantine system of "benefits" to choose and manage?
And compared to where? Germany or Estonia or Sweden or Spain? The complexities will vary wildly depending on the country (kind of like in the US, where lots of companies pick the state to base themselves in based on the combination of favourable laws and precedents and taxes).
there are certain sentences you can just tell would never be written by an American lol
California Consumer Privacy Act is a thing you need to take into account for Californian customers.
Illinois has a Biometric Privacy Act.
And who knows what Wyoming or South Dakota or Oregon have that you might take into account if your business falls under any of them.
most laws like CCPA also have some threshold where you already need to be pretty successful for it to apply to you.
for some select industries (biometrics & healthcare), yes you have a patchwork of laws.
Because they are plenty of companies just ignoring regulation and simply paying the fines, as the enforcement often a joke.
TBH, in this respect it's much the same in the EU.
Okay, what is the limiting factor? Because when I talk to EU business owners (admittedly, very few) - they point to lack of big EU capital markets, which is directly downstream of the policy environment. And when I talk to top EU human capital, they all point to the lack of competitive wages. There's a real difficulty in allocating capital to talented humans.
And, at least in Southern Europe, the income tax schedule is so aggressive it's hard to justify continuing working in many of these countries if you are highly talented.
Like, if you can tell me what the induced operator norm from l_2 -> l_2 is - probably you should come to the US and work at a biglab and make bank. What can you do in Portugal, Italy, Spain, etc.??
> Pieter Levels is at the end of the day an influencer, not a serious founder.
Sure, agreed.
I think it is a complete misreading to point to protectionism as the reason for Chinese success, but having a big unified domestic market for consumers along with massive saving rates and capital controls probably does help.
A few.
A big part is that the EU is a collection of countries that (with very few exceptions) have different languages and laws. For a company to serve Spain and France, for instance, it would need to translate everything, hire local lawyers and customer support agents. Considering the much smaller size of the countries (biggest one is 70 million vs 330 million in the US), the opportunity for "unlimited" growth is limited.
This also rebounds in the fact that when an American company makes it big, they have the resources to flood other EU markets and be cheaper/better than the local competition due to economies of scale and money based on their big successful US market. A French company making it big is still small compared to a US equivalent.
Then, there's the capital markets, no denying that. The money being thrown around the US is like nowhere else on the planet. Some of it definitely a bubble / unrealistic, but that doesn't matter. But in part it's because of the size of the total potential market that this is justified.
Education / national mythology also plays a part, I think (this is pure conjecture now). In the US, the "American Dream", "everyone can make it" etc is heavily ingrained. It propagates through the world with the help of Hollywood and other American cultural exports. In most EU countries, there isn't such a heavy emphasis on independence and "pulling yourself up by your bootstraps". "Hustle culture" isn't a thing. So for most people, it isn't something that comes naturally to them to start a company and work 100 hour weeks to be big and rich and successful and famous.
That's not to say there aren't such people, I went to 42 and have been to Station F and know some people in that universe. A decent proportion of my classmates wanted to make their startup and make it big, and some did end up starting their own companies.
Ding ding ding! When China does it with solar and EVs we call it "dumping". When Uber, OpenAI and Anthropic do it, that term is never ever used. VC funded US techs dumps harder than any Chinese industry ever has.
If you manage to get 10 million customers, your business is already successful on a gigantic scale, and you should have all the know-how in taking on the world. The success of other people is rarely the reason why you are failing in your own life. Start somewhere, do something.
> The money being thrown around the US is like nowhere else on the planet.
That's true and it's awesome. In Europe money is only thrown to real estate owners and any enterprising people with a dream are cordially invited to fucking forget about it, shut up, and fall back in line. Even if they already have a proven track record. They take their idea to the United States and are treated incredibly well in comparison. Even if their business will only be a niche business with limited reach, like 99% of businesses.
That's simply not true. Europe is a continent that includes startup powerhouse Sweden, UK that has a ton of them too, and France that is making massive strides (just check out Station F). It might be true in Slovakia or whatever, but you simply cannot say that broadly for the whole continent.
> That's true and it's awesome
Is it? It is to an extent, but it also encourages nonsense (Juicero, the AI spice mixer), pump and dump tech-isation of existing stuff with no business model (Wework) and outright scams (Theranos, Nikola).
If the barrier to entry to get money thrown at you is so low, a lot of it gets wasted. So the rest really really has to make it big, which gives some perverse incentives (like Uber dumping to driver old school taxis out of business to then jack up prices, because if they're not a monopoly/biggest fish in town, they wouldn't have been worth it)
Go search the web for startup financing in Sweden. One of the big banks has a page dedicated to this - but they're not offering their own money, instead they're pointing to government grants and government lending programs.
> If the barrier to entry to get money thrown at you is so low, a lot of it gets wasted. So the rest really really has to make it big
You can't do your accounting like that. One entity does not make up for the losses of another entity. Money is cheap in the US and thrown around to bad startups and good startups. In Europe money is expensive and mostly available to governments or for real estate feudalism.
Why work in the "europoor" countries when you can go to america and earn megabucks.
All of these purported EU-specific reasons completely ignore that things are the same elsewhere. It's the US that is the outlier.
That's not the reason the EU has no unified capital market.
The reason is that member states are reluctant to hand over control and power to EU. Just look at e.g. Eurobonds and the shitshow around that. Especially the rise of populist anti-EU parties all over, are causing this.
Ironically, the parties that "freedom lovers" like Musk are funding and pushing, are the ones causing the EU to not function as a single capital market and "not having a business friendly environment".
Capital controls are protectionist measures, but anyway, no.
> Okay, what is the limiting factor?
Let's look at which countries have a significant local software industry compared to population size.
- China
- US
- Korea
- You can argue for Japan and India but that's already starting to stretch.
- Yup, effectively no where else. Even in an "out of the way" place like Myanmar everyone uses Meta, with a nice little genocide to show for it. Sure, in Vietnam they use Zalo, and other places have a few other local players. But most of the famous US tech apps are dominant.
Is the EU the outlier here? No. Everywhere else US tech dominates. Meta, Netflix, Apple, Google, Uber, Spotify, Microsoft, Match Group, Paypal, Amazon, and on and on. They don't just dominate the EU, they dominate the world.
Except for the countries I named above, where at least some of the markets that US big tech competes in, instead have bigger local players. And even there, guess what?
Their market share is almost 1:1 linearly correlated to the degree of protectionism in those countries, all the way from China, then Korea, then India/Japan, and then everywhere else! Who woulda thought!
Why does Korea have much less US tech dominance than, say, Germany? Despite German companies theoretically having a big advantage: the German public is 100x more privacy conscious than the Korean one, and much less trusting of US companies.
I can tell you that it's not less regulations; Korea's GDPR is much more onerous than the EU's and so are investment regulations. On every single regulatory aspect, German software startups have it easier. But they were never protected. US tech was allowed to waltz in, dump their products - that's what they did, it's hilarious how now China "dumping" EVs and solar is suddenly an issue when it's exactly the strategy that US tech continues to this day; the AI companies are doing it right now! And the Korean companies were protected. Both by the rules burden, that local companies had to deal with too, along with intentional protectionism.
When it comes to solar and EVs, we all understand that a foreign country dumping their goods kills local industry. It's the exact same with software.
But then half of HN has millions on the bank exactly thanks to the above - this is where all those fat SV salaries have come from - so I do get the lack of desire to understand it.
Fundamentally BYD cars are cheaper because China has localized the complete supply Chain or has very good raw materials import and local refinement capability. BYD invested themselves and spend 20-25 years vertically integrating while European companies put out huge dividend and miss major technology trends.
The EV market in China was hyper competitive with a huge number of competitors and the subsidies were strategically phased out to turn the industry competitive rather then relaying on protection.
The idea that China has enough money do 'dump' products in literally ever sector that people accuse it of 'dumping' AND at the same time have enough money for massive infrastructure programs at home seems like coping to me.
> Uber is irrelevant in many places in the EU due to protectionism.
Thanks for another great example that proves the point :) Though I think there are markets where Uber Eats dominates even when Uber Taxi doesn't. Could be wrong.
Seems like you actually believe this. I think our starting points on reality are different enough that we are not going to have a productive conversation, I wish you and other Europeans the best of luck in your protectionism-led growth strategy. Make sure to not discuss it with any pesky macroeconomists who might lead you astray. take care
What it's terribly good at is adding burdens that the US giants don't face early on, slowing down the early growth between 28 fragmented markets. I don't know specifically about how China works, but the question is proving product-market fit, and for that, you need a lot of users fast.
In the EU, it's a different battle country to country as the media environment, the markets, the regulation etc. are all fractured.
Bingo, my friend. And good old american exceptionalism taught since the first days of school.
In the US, some ex-Googler might found a startup. Europe doesn't have the equivalent of FAANG. (Europe-wide companies are not quite as easy as US-wide)
Even if the super computer itself "fails", is the goal actually the secondary impacts to the economy?
(And in the US, we do our own fair share of picking winners / losers, especially in the current regime)
Cluster: for public benefit, cutting edge research in biotech, medical, robotics.
Levels: I want to create AI photos of people for my AI Slop startup
That's not what the quoted paragraph says and you can read the whole release if you want: https://ec.europa.eu/commission/presscorner/detail/en/ip_25_...
--- start quote ---
Apply AI Strategy
The Apply AI Strategy aims to harness AI's transformative potential by driving adoption of AI across strategic and public sectors including healthcare, pharmaceuticals, energy, mobility, manufacturing, construction, agri-food, defence, communications and culture. It will also support small and medium-sized enterprises (SMEs) with their specific needs and help Industries integrate AI into their operations.
--- end quote ---
I also quoted a paragraph from a document I will find when I'm not on mobile.
Levels literally wants to train AI Slop: https://x.com/levelsio/status/1981499900266193028
--- start quote ---
Train a foundational model for AI photos of people
--- end quote ---
My quote: Cluster: for public benefit, cutting edge research in biotech, medical, robotics.
Literal quote from your link: The Apply AI Strategy aims to harness AI's transformative potential by driving adoption of AI across strategic and public sectors including healthcare, pharmaceuticals, energy, mobility, manufacturing, construction, agri-food, defence, communications and culture.
You: your quote was misleading.
I'm sorry, I don't have the time or the patience with willfully ignorant and blind people getting their interpretations from AI slop engagement farmers.
Adieu
> I'm sorry, I don't have the time or the patience with willfully ignorant and blind people getting their interpretations from AI slop engagement farmers.
Riiight.
If these are public money, you want to reduce the blind grasping
> Slop has stumbled into accidental success on a number of occasions.
So, show me these occasions where AI slop led to "transformative potential by driving adoption of AI across strategic and public sectors including healthcare, pharmaceuticals, energy, mobility, manufacturing, construction, agri-food, defence, communications and culture."
Being honest with taxpayers about what research is is probably possible unless the population is low IQ.
And frankly, the dream scenario that Pieter describes where he somehow would qualify for these resources also wouldn't help kickstart the tech industry, and it's also not how it works in the states.
What does help, and what European governments (at least the one in The Netherlands that Pieter is from) actually do, is more funding for startups. If you're a startup founder in NL almost every angel you talk to has a matched funding deal with the government. That's such a smart way of keeping up with the US. Do you think US startups get free compute from the government? They don't even get subsidies most of the time. What they get is better funding because there's more capital available, and helping investors with that is exactly how you solve that.
Does government offering matched funding to investors actually help startups who are struggling to find (any) funding? If a startup can't find (any) funding, matching is irrelevant.
> Do you think US startups get free compute from the government? They don't even get subsidies most of the time. What they get is better funding because there's more capital available, and helping investors with that is exactly how you solve that.
Umm. I'm not really convinced that the political elites in Europe understand how to do any of this stuff well.
See also: https://www.eib.org/en/publications/online/all/the-scale-up-...
What’s worse, the parliament cannot originate law. Only the unelected Commission can do so. And they can do it behind closed doors. This is a setup that’s ripe for corruption.
However, they're appointed by the EU Council (the heads of state, most of them elected, some appointed by a national parliament), and approved by the (elected) European Parliament.
At the cost of some transparency, this does make it possible to select a bit more for management skills instead of just campaigning skills.
US presidential candidates are appointed, but they must then be elected by the public.
UK PMs must be elected MPs, or at least must face an immediate by-election (by constitutional convention) if not already an MP
If back-hand deals, internal political favouritism, nepotism or opportunism lead to the appointment of an EU Commissioner there’s nothing the public can do about it.
Is this really true? Aren't the outcome of primary elections much much more important in determining who will be the Republican and Democratic presidential nominees than elected officials and party officials are?
Specifically, an unreliable source that gives fast answers to questions says that a candidate who wins enough primaries to get over half the delegates is almost certain to be the presidential nominee for his or her party.
* Citizens vote in a nationwide election every four years.
* Those votes determine electors in each state — this is part of the Electoral College system.
* The Electoral College then formally elects the president based on those state results.
* The candidate who receives at least 270 of the 538 electoral votes becomes president.
So while there’s a layer of formality through the Electoral College, the president is ultimately chosen by voters, not appointed by any government body.
The same way that voters elect their national governments, which then appoint the Commission.
* It doesn’t have ongoing powers, authority, or political discretion like a parliament or the European Council.
* It’s a one-time, ceremonial body that meets once every four years to cast votes reflecting (in nearly all cases) the popular vote results in each state.
2. Electors don’t deliberate or choose freely.
* In almost every state, electors are legally bound (or at least politically bound) to vote according to their state’s popular vote.
* So the outcome is functionally determined by voters, not by an independent decision of the Electoral College.
3. The EU analogy misrepresents the role of the Electoral College.
* In the EU, citizens elect national governments, which then appoint members of the European Council, who negotiate and nominate the EU Commission — a genuine appointment process.
* In the U.S., the electors are not government officials making appointments; they are agents executing the people’s will as expressed in the election.
---
In the US, the people elect the President, because electors vote as their state’s voters directed.
The EU Commission analogy doesn’t fit, because that’s an appointment process made by governments, not a formalization of a direct vote.
Fact of the matter is, neither the US President nor the EU Commission are directly elected. Both are appointed with one layer of indirection between them and a direct vote.
Either we call both elected or we call both unelected. To do so for one, but not the other is anti-European propaganda.
It’s also enshrined in our respective governments and the foundational philosophies that underpin them. The US Declaration of Independence sets out to describe that the natural rights of the men who created the USA are preeminent and the Constitution lists how those rights may not be infringed upon, i.e. It creates laws that binds and limits the actions of government, something that was and has never been emulated since. Where across Europe, you simply do not have anything even remotely similar and the law inversely describes what you are permitted by government to do instead.
It is effectively descriptive vs prescriptive law and underlying philosophies. It is something I have had the hardest time on occasion getting my European friends to really internalize, seemingly because it’s so contrary to what their conditioned with all their life, i.e., the government is essentially the matter that grants you what it grants you, not that you have rights that the government may not infringe upon.
But to be fair, this possibly European tendency to dominate and control what you may and may not do and when and how, is and long has been creeping into the USA too, arguably since even the 12th amendment to the Constitution and getting worse with every amendment since, layers upon layers of contradicting and conflicting flaws and bugs that will be reading their ugly heed here in about two years, when Trump may run for President again. And if you don’t think he can, you simply don’t understand what a spaghetti code the Constitution is after the 11th amendment, hack after hack building up mountains of debt that is going to come due in our lives.
Ah the good old "Europe can't do Silicon Valley" trope.
Which I think many (the vast majority?) Europeans will fight you to the death to avoid admitting. The politicians just made some honest mistakes. No corruption at all. Just read the other replies in this discussion.
The EU is such a bizarre place because they treat capital and entrepreneurs with such massive distrust, but never really bothered getting rid of the quasi-static entrenched hierarchies from feudalism? Like I'll go to the UK or France and there will just be massive swathes of land owned by the nobility or 'former' nobility? Maybe start there but let your high-value human capital earn a good wage?
Those 1960s regulation have a 1000x larger effect then any land owners owning a bunch of unproductive farmland or highlands.
Not to say there are no issues with large land owners, but its nowhere near anywhere close to being the issue with housing prices and living cost.
Yeah, no, this isn't even remotely true.
The EU Parlament has it's issues, and it's atrocious how corrupt politicians like e.g. von der Laien fail upwards into leadership roles there, but they're both massively exaggerating the issues to a degree that it's hard to call the stated opinions anything but unhinged.
Its kinda like calling a Chihuahua "apex predator". completely out of touch with reality.
2. A credible scale effort for EU own silicon for AI Compute, wouldn't hurt either.
3. And this can only be achieved by vertical integration to combat fragmentation.
Would you prefer European AI sovereignty with 15% overhead costs from geographic distribution, or 100% dependence on Nvidia/OpenAI with zero European industrial base?
EuroAI: Europe’s Moonshot to AI Sovereignty
https://open.substack.com/pub/ifiwaspolitical/p/euroai-europ...
Yep, the US-government sponsored, open-weight LLM is miles ahead of EuroLLM
Just the base model and a template like "English: {text}\n{language}:" can also work with a bit of filter and retry logic
My experience with government funding is that they apply something and won't even try to sell it because selling is hard: you don't want to know that the thing you built is lacking nor that the competition is better. Especially the academic types don't. Yet I'm paying for these guys. Also, by funding the academics they won't even need to go to the job market.. But as I paid for their education I thought I was buying people who create value.
Perhaps the above is rather harsh and it's "not that bad", my subjective experience nevertheless.
Vaswani is an Indian born computer scientist, Shazeer is US, Parmar was born in India, Uszkoreit was born in Germany, Jones was born in the UK, Gomez is British-Canadian, Kaiser is a Polish computer scientist, and Polosukhin is Ukrainian.
Almost all of these people have PhDs and Master degrees. The ROI on academia is vast for society, including European universities. The thing the US does well is capitalize on that education, and sadly also try to steal credit for it as "American exceptionalism." If Europe and other countries learn how to keep their academics and get them working in local industries, America's edge will evaporate overnight.
The wider availability of capital is a bigger deal though. "Attention is all you need" is available to people on other continents to read, but a computer scientist in Europe that understood exactly how big transformers were going to be and why had less chance of funding than a webdev in California with a pitchdeck full of cliches and me-too GPT wrapper for an industry they'd barely touched does today.
To be clear, I don't oppose publicly funded education (nor immigrant academics, though I don't see how this relates?). What I do oppose EU trying to compete with tech giants as if they could - the incentives are not set up right, they won't succeed and the funds will be wasted.
There are a few variables here but at this point in time, private-funded innovation isn't different by much and all things considered, the difference isn't in its favor.
It seems like it, in most ways, it would be bad to train on 24 separate languages. That's just 24 partitions to the data. Seems really inefficient and better to simply train in the biggest (english) and translate.
I do think this will introduce some biases that correlate with the English language. It would be interesting to see more specifically what this means. But regardless, I don't think you can produce a competitive model with such a large subdivision of training data.
This model was released in 2024, and I couldn't find any links to the training data - is it just an open weights model?