That would only work for the percentage of the European population who uses English as their primary language. It'd be damned near useless.

I already noted that and plenty of languages are spoken outside the EU aside from just English.

>Google can then use non-private data to adjust the model to the EU market.

Google can do everything a EU-only company can do. They can also do things a EU-only company cannot do. The reverse is not true. So at worst they'll be roughly as good and at best much better.

And why is the company EU-only? What X-only company is trying to compete with Google anyways?

No they won't.

There are regional differences between the same language that can be pretty severe, all the way down to the pronoun level.

So, no they can't.

I don't think you understand the point. Google can get all the same data as an EU company. They literally can do the same job an an EU company by simply not leveraging US data. The fact they have US data doesn't mean they are forbidden of getting EU compliant data like any other company.

They can also get US data that doesn't follow EU privacy requirements. So Google has two source of data they can combine into a better product.

The same goes for the US. but there are always regional differences, the bigger point is that people are not just a simple lookup table, they can adapt to a personal assistant that only understands an imperfect understanding by changing they way they speak to the device. People will do this without even being conscious of it.

Nope.

In the US the dialects are very similar and usually separated mostly by accent.

When dealing with different countries, entire groups of words do not mean the same thing because either (a) the language has changed, or (b) those things literally did not exist at the time the split occurred. Training an AI on Parisian French will result in an utterly broken experience for Haitian French and Cajun French.

My point is, that if you trained an AI on Parisian French, people in Quebec would be annoyed but could adapt to the differences if they wanted to use the system. People in southern Germany may be annoyed they have to speak high German, but they know it. There are very few people, if any, who only understand Cajun French and the issue.

Also, the biggest difference would be the pronunciation differences vs words being different. The word differences are easily modeled after any ASR, if the developer understands the differences and plans for it in their NLU models.

Poverty wages, child labor, ignoring emissions regulations, evading taxes, etc are all ways a company could improve margins and charge lower prices. That doesn't mean you want to legalize all of that just to compete in the global market.

It's ultimately going to be a race to the bottom if you allow companies who engage in those practices all over the world into your market on even terms. That's honestly what tariffs are for. Tax your citizens for engaging in commerce with companies that gain efficiencies through operating in ways that disagree with local values.

Couldn’t you use the same line of argument for tariffs on a vast number of imports? (Electronics, software, services, etc.)

It could even extend to most consumer product. e.g. Some obscure chemical compound for plastics is much more costly to produce in an environmentally friendly way, therefore it’s nearly all produced in the least regulated jurisdictions, therefore the compound and all derivative products have to be tariffed from those jurisdictions?

Why do you put words in my mouth that we should legalize something when I said nothing of the kind? If this is how you react when someone points out flaws with a legislative approach then good luck ever getting one that isn't flawed.