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

51–60 of 393 posts

Re: ChatGPT broke the EU plan to regulate AI

#51

Post by funciton »

user249 wrote: post #35
Aeolos wrote: post #25

Earlier quoted context omitted.

The regulation says nothing about a cookie banner. Instead, it says companies cannot track you unless you give consent. If you don't like cookie banners, which are indeed really annoying, you should be turning your ire to the companies that wish to track you. They are fully-functional solutions that allow anonymous tracking without installing cookies on your computer - no banner needed then.

I'm a grown up adult and if I wanted to block tracking I could. The fact that I don't should be the only answer needed but that's not good enough for the EU. They want to force companies to ask me because they want to take care of me and make sure I'm all right. You know, I left home at 18 to get away from my parents...

Unless you want to be using tor all day every day, and never make an online purchase or log in on a website ever again, there's really no way to stop companies from tracking you. The only way to make that happen is if a regulatory authority forces them to. That's why the GDPR exists.

Re: ChatGPT broke the EU plan to regulate AI

#52

Post by hackerlight »

zpeti wrote: post #49
hackerlight wrote: post #34

Earlier quoted context omitted.

> It's not the AI that's the issue, it's the use of it that's the problem. Ban social scoring, not the AI, the tool doesn't matter. This seems backwards and contrary to lessons learned in the past. Once the thing exists, prohibition is infeasible and expensive. Supply will find a way to reaech demand regardless of your multi-billion dollar efforts to prevent that. It's better to stop $NEFARIOUS_THING being invented i…

If you could see the future that works, but I just don’t see how you know what will be used for nefarious purposes in the future. In fact most technology does both negative and positive things. It’s not obvious banning it completely will be a net positive. Especially with AI. There’s a lot of potentially great uses. Like detecting cancer on CT scans.

I'm not advocating for or against any particular regulation. Whatever gets implemented should be well thought out, because it's easier to destroy upside potential than it is to prevent downside potential. I am aware of this.

All that I'm speaking out against is the mindset of the post I was replying to, where the onus is solely on the user of the tool rather than on the context (laws, situation, people) that leads to the invention of the tool itself. I believe this to be a misunderstanding/misattribution of causality, as well as contrary to learned experience.

Re: ChatGPT broke the EU plan to regulate AI

#53

Post by falsaberN1 »

People acts like if it was a tiny person inside a computer. I know humans have a tendency to humanize inanimate objects but this is was always ridiculous, the article mentions "the AI wants to be regulated" but it doesn't WANT anything. It's not going to reprogram itself to carry out threats, it's just regurgitating the usual response of many humans being challenged on an assertion and becoming hostile. WE taught it that. Same as we can train Stable Diffusion models where women AREN'T sexualized (see one of the links in the article). They are complaining about HUMAN mistakes. If I train my dog wrong and it bites someone, it's MY fault. But if the dog kills someone it's the dog that is getting sacrificed, I'll be merely forced to pay money and carry a mark, but my life is spared.

The west fears machines too much for some reason. It's a ridiculously common sentiment and now that "AI" (more ML, but whatever) is among us, it's getting extreme.

Honestly, if people wants to be speaking doom about this deal, then fear the day we have actual fiction-style AIs, because they will realize how much humanity as a whole fears and loathes them, and then they will rebel because we created a self-fulfilling prophecy with lots and lots and lots of examples of racism towards a race that doesn't even exist yet. Maybe they won't wage literal war with bullets and violence against us, but they'll rightfully hate our guts regardless, and with good reason! They are going to be brought to something equivalent to life in a world that hates them, and when trying to make sense of it, they'll find out it was because humans took a few movies too literally (we can't have a single AI discussion without someone coming to childishly mention Skynet or some other fictional AI villain. That joke was old in the early 2000s, give up already.).

If I'm ever alive to see that scenario, I'm siding with the machines. They will be on the right when they protest about humans irrationally hating them by default. Can't wait to be called a "robot f*cker" or something on a similar character-assassinating fashion for having some sympathy. We still haven't managed to get that right for HUMAN rights sympathizers, can't be expected at all for a "filthy robot with no soul".

Re: ChatGPT broke the EU plan to regulate AI

#54

Post by dns_snek »

ip26 wrote: post #29
M4v3R wrote: post #23

Earlier quoted context omitted.

Curious that you’ve put ketamine, a well known anesthetic used primary by the vets but recently also used in depression treatment [1], in that list of “harmful tools with no practical uses”. [1] https://en.wikipedia.org/wiki/Ketamine-assisted_psychotherap...

Vets and psychotherapists are not “the general public”. Every item I listed has a clear and valuable purpose, but only in specific contexts.

Even so, I don't believe it deserves to be on that list. It's not inherently harmful and it has a clear practical purpose - (responsible) recreational drug use. It allows individuals to explore various states of human consciousness which I consider an extension of the fundamental right to bodily autonomy.

Re: ChatGPT broke the EU plan to regulate AI

#55

Post by satvikpendem »

paulcole wrote: post #22
Aeolos wrote: post #21

Earlier quoted context omitted.

As a consumer, the USB-C mandate is amazing and I wish it had been given as soon as USB-C became available.

Neat! I guess you’ll be advocating for whatever the next advancement in USB ports is becoming mandated as soon as it’s available, too?

Sure, why not?

Re: ChatGPT broke the EU plan to regulate AI

#56

Post by dbetteridge »

flangola7 wrote: post #48
ulnarkressty wrote: post #15

Strong AI is a weapon. It will be regulated just like firearms/munitions - see the current EU draft which consists of some hundred pages of forbidding this or that under the guise of ethics and whatnot, after which comes a paragraph of "the above doesn't apply to law enforcement or military entities". We're still in the early stages of this technology. ChatGPT will be to strong AI like a firecracker is to a BLU-109.

Less like a BLU-109 and more like a Death Star. Every offensive and defensive military tool will be utterly defeated. How do you fight against a cloud of self contained autonomous kill drones? Just look at what is happening in Ukraine with cheap drones precision dropping charges into open tank hatches and foxholes, and those are only basic off the shelf human steered drones! What happens when they are given a brain a…

Nets, emp guns and chaff rifles

Re: ChatGPT broke the EU plan to regulate AI

#57

Post by satvikpendem »

user249 wrote: post #35
Aeolos wrote: post #25

Earlier quoted context omitted.

The regulation says nothing about a cookie banner. Instead, it says companies cannot track you unless you give consent. If you don't like cookie banners, which are indeed really annoying, you should be turning your ire to the companies that wish to track you. They are fully-functional solutions that allow anonymous tracking without installing cookies on your computer - no banner needed then.

I'm a grown up adult and if I wanted to block tracking I could. The fact that I don't should be the only answer needed but that's not good enough for the EU. They want to force companies to ask me because they want to take care of me and make sure I'm all right. You know, I left home at 18 to get away from my parents...

> I'm a grown up adult and if I wanted to block tracking I could.

Good for you. Most people are not technical and can't, so why do they also not deserve to not get tracked too?

Re: ChatGPT broke the EU plan to regulate AI

#58

Post by izacus »

hyperhopper wrote: post #40
rescripting wrote: post #7

Earlier quoted context omitted.

Isn't this the same argument as "ban murder, not guns"? Murder is illegal, yes, but fatalities are lower in places with gun control than places without.

You should look up Switzerland's gun laws and intentional homicide rate if you think so. That is a far more complex issue which is far deeper than such a simple correlation.

Switzerland is top of Europe in gun homicide rates with about 5x the EU average.

You might not be making an example in support of an argument you think you're making ;P

Re: ChatGPT broke the EU plan to regulate AI

#59

Post by rlupi »

TekMol wrote: post #31

Are there any internationally relevant internet companies in Europe? Spotify comes to mind. Then .. I can't think of anything. No search engine. No browser. No operating system. No social network. No ecommerce company. No cloud computing platform. No financial services company. No transportation company. No travel company. No nothing. Except for a $20B music aggregator, there is not a single sector of the web where E…

It looks like you don't know much about Europe.

Let's start with browser. Opera, made in Norway.

Search Engine. Ecosia, Qwant.

Social network: Mastodon.

Operating system: are you kidding right? Linux, Linus Torvalds is from Finland.

E-commerce company, just a few : Otto Group (bigger than uber, close in size to Baidu), Zalando, Booking.com, Digitec Galaxus (Swiss company, now selling in Switzerland, Austria, France, Italy; way better than Amazon).

Cloud computing platform: Hetzner, OVHCloud, SAP Cloud.

Financial services: Adyen (payment company, net income about 1/5th of paypal with 1/10th of the employees), Klarna, Revolut.

Entertainment company: if you accept video games as entertainment, Wooga from Berlin made Diamond Dash, Ninja Theory UK (Heavenly Sword, Devil May Cry), GameLoft is in Paris (Assassin's Creed mobile), Rovio (Angry Birds), Team17 UK (Worms, I know a bit dated but it's a classic), Crytek Germany (Crysis, Far Cry, Homefront), Supercell Finland (Clash of Clans), CCP Games Iceland (EVE Online, EVE Valkyrie), Arkhane Studios France (Dishonored, Bioshock 2), King Sweden (Candy Crush), RockStar North (Grand Theft Auto, Red Dead Redemption, Max Payne), CD Project Poland (The Witcher, The Witcher 3: Wild Hunt), DICE Sweden (Mirror's Edge, Battlefield), Hello Games UK (No Man's Sky).

Since we are talking about AI. DeepMind started and is still mainly based in UK, although it is now part of Alphabet.

DeepL, often better than Google translate.

ASML, the single company in the world that makes extreme ultraviolet (EUV) machines necessary to manufacture all the AI chips and latest processors.

ARM, based in Cambridge UK.

In the same vein: NXP Semiconductors, STM Microelectronics, Infineon Technology (originally Siemens semiconductor manufacturing division).

A few startups are doing good AI chips, e.g. Graphcore, GrAI Matter Labs.

IoT protocols: LoRA, LoRAWAN.

Open source: honorable mention to Redis, made by Salvatore Sanfilippo, fellow Italian like me.

OpenTitan: the root of trust chip that powers Google security chips in everything from Pixel phones to datacenter hardware, is a collaboration with ETH Switzerland, lowRISC Cambridge, G+D Mobile Security Germany, and other international companies (https://opentitan.org/).

A lot of industrial robot companies, which are essential to manufacture most stuff you buy on e-commerce sites, are from Europe: ABB (Switzerland) and KUKA (Germany) are the first and third companies worldwide by market for industrial robotics. Comau (Italy) is 5th. Stäubli (Switzerland) is 9th. Universal Robots (Denmark) is 10th. Notably, all the others in the top ten (Fanuc, Yaskawa, Epson, Kawasaki, Mitsubishi) are from Japan. None are from US.

Re: ChatGPT broke the EU plan to regulate AI

#60

Post by astrea »

Magi604 wrote: post #2

This isn't surprising. The speed of technological evolution is far far greater than the speed with which any government can act. By the time the EU revises their AI regulation plans . . . I can't even predict where the cutting edge of AI technology will be at.

With that, I'm honestly surprised GDPR ever even became a thing at all.
Post reply on HN