Hmm. I think I understand where you are going but given that LLMs appear to need injected randomness to work effectively, determinism isn't really on the table so we kind of stop talking about the LLMs we use when we (strictly speaking) consider the case where they could generate deterministic sequences?
Not that this in any way takes away anything from your argument that "it's that language is imprecise". That still holds true, but it is on the input side.
Why would LLMs need injected randomness to work effectively? Their nondeterminism isn't by design, it's largely a byproduct of how (batched) inference actually happens on hardware [0]. In case you are referring to the actual token sampling (be it temperature, Top-K or whatever), i wouldn't really see this as part of the LLM itself (And I've frequently seen benchmark sites run them at t=0 anyways)
It is hypocrisy because Meta are knowingly creating a product that is perfectly suited for filming or photographing people without their knowledge. I find that it speaks ill of someone's character to pretend otherwise. And then they react badly when the very thing that these products enable happens to them. Demonstrating how this is a problem. They try to pretend it isn't a problem and then their actions demonstrate that it IS a problem.
It isn't because we do have laws that actually cover these things already. Of course, they're not the same across the world, law enforcement are not always aware of what the laws are, and this causes confrontations and sometimes overreach. There are also different legal systems that interpret laws differently. For instance in some legal systems the letter of the law carries more weight. In other legal systems the intent of the law is what matters.
There are two great risks here:
1) we pretend we can get by with existing laws. That the existing laws had enough foresight to meaningfully fulfil what the laws were intended to accomplish.
2) we can overreact and damage legitimate freedoms in our eagerness to solve an immediate problem.
For instance, where I live there is a law on the books that prohibits photographing someone being arrested. The law was well-meaning. It was there to protect people from being plastered all over the media until guilt has been determined in a court of law. But it would also prohibit someone from documenting police brutality during an arrest. So the law was never enacted because it failed to balance interests properly.
Right now, where I live, there is a real risk that rules about photography may become overly strict because politicians feel they have to demonstrate an ability to act. So they may end up passing laws that have unforeseen consequences while probably failing to make much of a difference anyway.
We know why these cameras pose a problem. Let's not insult each other's intelligence by pretending they are exactly like poking a traditional camera in someone's face. They're not and we should be able to agree on simple facts. But we also know that we need to have a public discussion about what we're going to do about this so we don't end up introducing draconian laws that take away freedoms we should have. For instance the freedom to document abuses of power, criminal activity, artistic expression etc.
It's not "perfectly suited", it's EXPLICITLY DESIGNED TO do that. And it's very well designed and very effective.
But hey it's OK because Meta only want to record everything to train their AI, who cares if some of their friends use it to create CSAM material, their third-world country workers upload it to PornHub and the NSA secretly pipe all of that through their servers (or servers managed by MI5 in the case of Americans).
They added a recording light and made covering it stop recording or photography. They went through extra effort to at least try and provide a public signal someone is recording.
Depending on your environment (outdoors/bright sunlight and angle to the wearer) it can be pretty difficult to notice the light. And not everyone knows what the light is if they even notice it.
That functionality that is supposed to stop the recording if the light is covered is not perfect. You can find lots of "hacks" being offered online and apparently some of them work.
It's pretty obvious Meta did the bare minimum here. It could have made these much more conspicuous or decided, wisely, not to unleash this crap on society given the privacy and abuse concerns that clearly outweigh the benefits.
You can conclude by the fact that they hid a recording device into what looks like a normal pair of glasses as proof that the idea is to allow people to look like they're recording people without looking like they're recording people.
They could have made it much more effective - better optics, better microphones, better battery - but that would have made it extremely obvious.
I think Canada might have an opportunity to help Europe at least reason a bit about regulations.
For the most part I am thankful that the EU regulates things strictly. As long as I am the consumer of what is regulated and not the maker. Because while the regulation is well meaning, it is not always effective or efficient. And sometimes it achieves absolutely no positive results.
And yet, I am generally for regulating a lot of things.
For instance the cookie-popup-horseshit is failed regulation that came about mainly because regulators were squeamish about real solutions that would have required constant judgement calls. Instead of posing an existential risk to companies that run roughshod over our privacy, they created impotent regulation that is a dual failure: it protects exactly nobody and it wastes everyone's time multiple times over.
Companies that do bad things should face existential risks. It is the only mechanism that works. Fines only work for small players. For large players it is surprisingly often a calculation where the numbers dictate you DO bad things because the downside risk is mostly quantifiable and financially acceptable.
Yes, sanctioning companies guilty of egregious violations of privacy would have been noisy, but it would have been the right thing to do. And it would have been cheaper and more effective than what we have today.
Quality was the norm when people had little or no disposable income and expensive things had to last for a lifetime, and some less expensive things (like shoes) had to last years.
Quality was what you judged products on and the origin of brands.
The overwhelming majority of brands today are meaningless because they have nothing to do with quality. They are either status markers or tools for exploiting our tendency to form habits.
You need go no longer than your local hardware store. Almost all the consumer grade electrical tool brands today are from perhaps 2 manufacturers. But you are likely to have a preference that you tell yourself is grounded in a rational, informed choice. Except, it mostly isn't. You probably buy the same toothpaste every time. Not because you have done an exhaustive study of toothpaste and concluded it is the best one for you. But because it is what you are used to buying.
(Yes, there are still some professional tool brands that make their own electric tools, but the kind of stores consumers shop in don't have those since they cost an arm and a leg)
Yes, quality used to matter. A lot. But most people alive today have not lived in those times. People have a hard time imagining things that are outside their own experience. So much so that they can even conclude "it can't be if I haven't experienced it".
I am starting to see how many developers actually need to re-learn programming in the age of LLMs.
A while back Claude went down in the middle of a somewhat frantic initial deployment of a product to production at a company where a friend of mine works. And suddenly nobody was able to do anything. Because nobody had actually read the code and had no idea how it worked.
So essentially: much of their day to day work now depends entirely on the availability of a couple of frontier LLMs.
If, for a moment, we set aside the fact that Europe is strongly incentivised to depend less on the US, and US companies, it should worry us that so many sectors are dominated by 1-3 or so players. One of the selling points for market capitalism, which we at least pretend to believe in, is efficient allocation of resources through competition. But when the market is dominated by de facto monopolies there exists no real competition and there is no real market.
Even in the previous political climate this should, objectively, be deemed a problem. Having 90% of all eggs in one basket is neither good, nor does it constitute functioning market.
I'm not sure what the Cloudflare market share is in the US, but if it is the same as in Europe, this is a problem for US companies as well.
How would one hold Google accountable? By fining them $1bn? $10bn? $100bn? To them it is just the cost of doing business.
We're beyond holding huge companies accountable. The only way they could be held accountable is if there was personal risk involved for the CEOs. There isn't.
At this point, breaking them up into smaller companies. Then start fining the individual companies so the balance sheets feel the fine more. Increase the fines exponentially for each violation.
What is the likelihood of that happening? Realistically? Does anyone think government can, or even wants to, break up large companies? The large companies _own_ the politicians.
If I am the evil corporations owning the state I would love to make all citizens pathetic cynics who think they have no capability for change and therefore bears no responsibility.
If a scam ad is found, at the very least, require them to disclose business records to show how much money the scam ad made (revenue, not profit), and make them repay it. Leave it up to the platform whether to recoup the paid-out share from the publisher or not.
If they are caught doing it repeatedly without taking adequate measures to stop it, treat them as an accessory to the crime.