Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ChatGPT broke the EU plan to regulate AI

politico.eu

Post reply on HN
Print view View original

371–380 of 393 posts

Re: ChatGPT broke the EU plan to regulate AI

#371

Post by anigbrowl »

GuB-42 wrote: post #268
anigbrowl wrote: post #91

Earlier quoted context omitted.

What an absurd argument. Social scoring doesn't get computer by hand on paper because it isn't efficient to do so. Introduce automation and it suddenly becomes practical. Same with mass surveillance and a bunch of other stuff, eg scams: https://news.ycombinator.com/item?id=35033971 the amplification factor works for the good and for the bad, so you need to focus just on the bad That's just hand-waving. You have no pl…

You absolutely don't need a computer to do social scoring. Computers let you take more parameters into account, and do more complex calculations, but you can do basic social scoring by hand, we have been doing that since civilization existed, in a more or less formal way. Traditionally through ancestry (and by extension, race). And the way we prevent abuse is with anti-discrimination laws, and AI won't change the dea…

And the way we prevent abuse is with anti-discrimination laws

Have you not noticed how heavily those are contested in many jurisdictions. If you discover some way to draft legislation on morality such that it's effective, consistent, and widely accepted in a disparate population, I look forward to hearing about it. I can't wait to read about the solution to spam, scams, and discrimination deployed at industrial scale. In the meantime, it's wise to take the amplifying aspect of technology into account and consider its potential for abuse as well as for good.

Re: ChatGPT broke the EU plan to regulate AI

#372

Post by everforward »

mirchibajji wrote: post #321
aftbit wrote: post #11

Earlier quoted context omitted.

Yes, which is also a valid argument. Tools have no alignment. The people using them do. Thus, one should regulate the people, not the tools.

This is a case of people actively building and deciding to use a tool with plausible deniability built-in. How would you regulate people when they shirk any accountability? It is much more reasonable to require that people use a tool that shows its work.

> How would you regulate people when they shirk any accountability?

This happens with or without AI. It's not too hard to make an algorithm that discriminates against people without doing so explicitly. You can use certain bits of data as proxies for whatever characteristic you want to discriminate against (e.g. ZIP code, household income, marital status, etc).

Due to the myriad of ways to arrive at a discriminatory conclusion, it's easier to regulate results than tools.

One potential solution to your question is to make the laws carry strict liability (I'm assuming the EU has something along those lines). Plausible deniability no longer exists because intent doesn't matter. The company is liable if someone can demonstrate discrimination, regardless of whether it was intentional or accidental.

That ends up pushing towards something similar to what you want. It encourages a tool that can show its work to fend off lawsuits without being directly tied to the tool itself. The other alternative is extensive testing to make sure discrimination doesn't happen, but I think that will still be worrying to companies due to the inability to prove a negative.

Re: ChatGPT broke the EU plan to regulate AI

#373

Post by hyperman1 »

akomtu wrote: post #114
talideon wrote: post #70

Earlier quoted context omitted.

For the curious, in Switzerland, you don't keep ammo at home, there's proper registration of weapons with proper background checks, and it's treated more as a means of national defense, so having lots of guns is going to get you odd looks from people. Guns aren't seen as something for personal defense: that's the job of the police. In other words, Switzerland has a much healthier attitude towards gun ownership and us…

Switzerland has a more traditional, not healthier, relationship between the people and the state: the people are subjects who are given tiny freedoms, but are kept on tight leash, because the state has a deep distrust of its people. The abberation with guns is just insecurity of the state: it's small and needs its people to act as part-time military in reserve. If it had a bigger military, the fake gun rights would d…

They are one of the most people-empowering nations ever. If enough people indicate it(50k signatures in 100 days cfr google), a legally binding regerendum must be held. A simple majority can then change the law.

I wish more nations could have that kind of direct acces to their democracy. Even so, the majority of these referenda fail to change anything, so presumably people tend to make more noise than wanting actual change.

Re: ChatGPT broke the EU plan to regulate AI

#374

Post by cccbbbaaa »

anothernewdude wrote: post #121
cccbbbaaa wrote: post #85

Earlier quoted context omitted.

Neither ePrivacy nor GDPR require cookie consent popups, instead they basically say that tracking requires consent. GDPR only mentions cookies once, in its recitals. It's the adtech industry that decided to ignore DNT.

Correct. Cookie banners were required before GDPR. GDPR is the one that makes dealing with EU citizens online a risky enterprise, so the large entrenched players have the advantage.

I know that cookie banners exist since ePrivacy. That's why I mentioned it. But ePrivacy does _not_ mandate them. The adtech industry could have respected the DNT header.

Re: ChatGPT broke the EU plan to regulate AI

#375

Post by cccbbbaaa »

IMTDb wrote: post #364
cccbbbaaa wrote: post #85

Earlier quoted context omitted.

Neither ePrivacy nor GDPR require cookie consent popups, instead they basically say that tracking requires consent. GDPR only mentions cookies once, in its recitals. It's the adtech industry that decided to ignore DNT.

> Neither ePrivacy nor GDPR require cookie consent popups Except that it does. The law specifically prohibits any form of consent that is not informed and specific. As a consequence, a user cannot just consent - or disallow - cookies globally. He has to tick a box for every single domain on earth; and can only do so after reading the specific information box associated to said domain. As a user, saying "I am OK with…

I'm afraid you are wrong for the global disallow button, as informed and specific consent is necessary to allow the processing of personal information, not to forbid it. See the definition in article 4(11), and the definition of the basis in article 6(1)a.

Re: ChatGPT broke the EU plan to regulate AI

#376

Post by dang »

dmitriid wrote: post #230
worrycue wrote: post #182

Earlier quoted context omitted.

I think the categories should be: - Accept all - Mandatory for function non-tracking cookies only - Reject all There should be a standardised browser accessible interface so browsers can automatically choose the one you want on your behalf based on your browser settings.

[flagged]

We've banned this account for repeatedly breaking the site guidelines and ignoring our many requests to stop. Not cool.

I appreciate your good comments but your frequent disregard for the intended spirit of this place causes more damage than your good comments add goodness, and set an unacceptable example for others.

https://news.ycombinator.com/newsguidelines.html

https://news.ycombinator.com/item?id=31682147 (June 2022)

https://news.ycombinator.com/item?id=31114217 (April 2022)

https://news.ycombinator.com/item?id=29913069 (Jan 2022)

https://news.ycombinator.com/item?id=25320264 (Dec 2020)

https://news.ycombinator.com/item?id=24319283 (Aug 2020)

https://news.ycombinator.com/item?id=21718992 (Dec 2019)

https://news.ycombinator.com/item?id=19814870 (May 2019)

https://news.ycombinator.com/item?id=16547831 (March 2018)

Re: ChatGPT broke the EU plan to regulate AI

#377

Post by narrator »

roca wrote: post #311
narrator wrote: post #111

Earlier quoted context omitted.

You have your drones. You also have your own personal chatbot that tells you when to ignore instructions from super-persuader AIs that would get you to kill yourself , spend all your money, or fall madly in love with an LLM if you didn't have a personal chatbot to protect you from their persuasive capabilities.

What if your personal chatbot was subverted by the super-persuader, or was controlled by it from the beginning? Impossible to be sure. We're going to have to stop talking to machines.

Your personal chatbot to protect you will at first be a luxury and then be more of a commodity. It will run on your own machine. Right now you can run the big Facebook llama model on a single machine if you buy a $15,000 A100 with 80GB of VRAM. Churches, political parties and other moral associations will fine tune these chatbots based on their values.

Re: ChatGPT broke the EU plan to regulate AI

#378

Post by int_19h »

anigbrowl wrote: post #370
int_19h wrote: post #156

Earlier quoted context omitted.

And those websites that are designed for EU users will have EU rules applied. It's just that it'll remain trivial to go to the other website which does not have that, and people will do this as long as that other website feels more useful to them, for whatever reason.

As I said above, the widespread compliance with GDPR opt-out regulations refutes your point. Perhaps there are some people boycotting all sites which offer such things, but I think their numbers are negligible.

The popups are cheap enough to implement, and don't meaningfully affect functionality of the site once you get past them; they're, at most, a minor annoyance.

This scenario is different, though, if websites with non-EU-compliant AI backend can offer additional features. Which is practically a given.

Re: ChatGPT broke the EU plan to regulate AI

#379

Post by int_19h »

talideon wrote: post #357
int_19h wrote: post #115

Earlier quoted context omitted.

Czechia is a European country you want to look at for an example that you can have gun laws that allow civilians to own handguns and carry them for defensive purposes without problems. US is somewhat unique in the Western world across the board - no public healthcare, generally poor social safety net etc - which means that exercises in causal linking of those unique traits to outcomes are wild guesswork, if one is ta…

First up, Czechia has proper licensing and strict background checks. Maybe not the best choice to compare to the US. I don't have to take a wild guess around this kind of thing: I'm specifically referring to the gun culture of the US and comparing it to that of Switzerland. There's a different attitude to gun ownership in the two countries, one of which places strong emphasis on responsible ownership and the other wh…

Sure, and I didn't claim that Czechia is the same as US in this regard. I'm merely pointing out that you can have a country with a gun culture that, among other things, enables personal self-defense without all the associated problems that US has which people often assume are intrinsic to gun ownership and defensive use.

I disagree that the situation in US has nothing to do with healthcare and social safety net, though. For example, suicides constitute most gun deaths in general, but this is worse in US because of its generally poor situation with preventative health care, especially mental health care. Similarly, poor social safety net means a lot more people ending up in poverty, which drives up crime including violent gang clashes that cause the majority of gun-related deaths in US. This isn't to say that the prevailing attitudes among American gun owners do not contribute to the problem, either - but I'm sure that, if US would catch up with e.g. Canada when it comes to taking care of its own population, the numbers would strongly reflect that.

And yes, it is ironic and sad that support for RKBA is strongest on the right, which is also the cause for all this. It's not exclusively them, though. The running joke on the subject in many hard left spaces is that "if you go far enough left, you get your gun rights back".

Re: ChatGPT broke the EU plan to regulate AI

#380

Post by int_19h »

dmitriid wrote: post #225
int_19h wrote: post #115

Earlier quoted context omitted.

Czechia is a European country you want to look at for an example that you can have gun laws that allow civilians to own handguns and carry them for defensive purposes without problems. US is somewhat unique in the Western world across the board - no public healthcare, generally poor social safety net etc - which means that exercises in causal linking of those unique traits to outcomes are wild guesswork, if one is ta…

> Czechia is a European country you want to look at for an example that you can have gun laws that allow civilians to own handguns and carry them for defensive purposes without problems. Let's see: --- start quote --- A gun in the Czech Republic is available to anybody subject to acquiring a firearms license. Gun licenses may be obtained in a way similar to a driving license – by passing a gun proficiency exam, medic…

(I don't know why your comment got flagged, since it's factual. Whoever did this - please don't, this is not a feature meant to be used indicate your disagreement.)

I'm not claiming that Czech gun laws are identical to US. I'm claiming that you can have shall-issue concealed carry and broad access to "assault weapons" for civilians without gun violence problems as seen in US.

Post reply on HN