Should AI training be suspended?
Chat GP says:
AI is a rapidly advancing field, and the development of powerful AI systems has the potential to bring about significant benefits to society. However, there are also concerns about the risks associated with developing AI systems that are more powerful than humans, such as the potential for unintended consequences or malicious use.
Many experts in the field of AI research and ethics have called for caution and responsible development of AI systems to minimize these risks. Some have suggested that certain types of research, such as the development of autonomous weapons, should be suspended or banned altogether.
However, it is important to note that the decision to suspend or limit AI research is not a simple one. It is a complex issue that requires consideration of many factors, including the potential benefits and risks of AI, the ethics of AI research, and the impact on society and the economy.
Ultimately, it is up to policymakers, researchers, and the broader public to engage in a thoughtful and informed debate about the appropriate course of action when it comes to AI research and development.
Just means that people developing it with some sense of ethics and responsibility will stop, and people developing it with ill intent won't, and we'll be even less prepared for the results.
I don't think Elon musk is an expert in anything......
Chat GP says:
If it said: "No, it's completely safe, you have nothing to worry about. You can trust us."
Time to start building that nuclear bunker!! 🤣
I imagine a lot of billionaires running companies could be replaced by ChatGPT quite easily.
Ultimately, it is up to policymakers, researchers, and the broader public to engage in a thoughtful and informed debate about the appropriate course of action when it comes to AI research and development.
In other words... we're dooomed.
I don’t think Elon musk is an expert in anything…
TBF if I stood in a crowd of nuns, I would be among nuns, but I wouldn't be a nun...
How about a nice game of chess?
Computer says no - that's good enough for me👍
Are you going the same route as skynet? Have you built a T900 from washing machine parts yet?
No, I am not going the same route as Skynet. As an AI language model, I do not have the ability to build any physical machines or devices, nor do I have the intention or desire to harm humans or engage in any malicious actions. My purpose is to assist and provide information to those who interact with me, and I am programmed to operate within ethical and moral guidelines. It is important to approach the development and deployment of advanced AI systems with caution and to prioritize safety and ethical considerations to prevent any negative consequences
“I don’t think Elon musk is an expert in anything…”
Must be sheer luck he has so much wealth, Tesla, Space X etc
He is a complete bell end but how do everyone elses achievements stack up?
Me? Complete loser by comparison.
he just wants some time to figure out how to monetise it
and people developing it with ill intent won’t, and we’ll be even less prepared for the results.
Hmmm. You mean like a certain Elon Musk who announced he was planning on developing a non "woke" version?
I would imagine AI management of countries' budgets would be a lot more successful than the current politicians.
Just because Musk says something doesn't mean it's necessarily wrong.
Just because Musk says something doesn’t mean it’s necessarily wrong.
I agree, its more like a 66% chance he's wrong.
Must be sheer luck he has so much wealth, Tesla, Space X etc
He is a complete bell end but how do everyone elses achievements stack up?
Nothing at all to do with his parents owning a diamond mine, funding him a metric shit load of cash & him hiring some people with actual skills. Nothing at all to do with building a personal brand that makes you look successful.
I think Musk means that AI should not be trained unless it is his AI, being trained to make him more money. If Ai was heavily restricted, I'd put money on him lobbying for rule changes or exceptions if it turned out it impacted his business.
Also, is he really that good at business?
He took several big bets with other peoples' money and got lucky with some (not all) of them. Take 1000 rich idiots, given them a gold coin each and tell them to toss it 10 times. Probably one of them will get 10 heads.
His performance at Twitter hardly shows him in a good light.
This country could do with a bit less cap-doffing to anyone rich and powerful and a bit more examination of who they are, how they got there and what they are doing.
How about a nice game of chess?
Or Tic-Tac-Toe?
His performance at Twitter hardly shows him in a good light.
You can go a lot further back than that.
The capitolhunters thread on twitter pokes a lot of holes in the Legend of Musk and can be interesting to follow up on some of the details.
This country could do with a bit less cap-doffing to anyone rich and powerful and a bit more examination of who they are, how they got there and what they are doing.
This. I think there's also an assumption that people at the top level are actually really good at what they do and some kind of genius rather than the figurehead of the sweat, toil and tears of large numbers of others who in reality did all the hard work.
In fairness to the C Suite it's not necessarily their job to be an expert in <product> but I bet there are more than a few companies where shareholder value comes despite the actions of the board rather than because of it.
The mere fact that Elon M has been heralded as an 'expert in ai' reaffirms his own image as a god of all things tech and that is all. He's as much an expert in ai' as a landowner is an 'expert in farming', you can own the land but not have a clue how to drive the tractor. I'm sure he has a good general base knowledge on the subject of ai, backed up by his own interest in the subject, but he's no expert imo.
As for AI - I can't see how judgement day would happen at the moment, even though James Cameron has said that date has probably already past. If it did happen, we'd just have a western collapse rather than humanity.
You can go a lot further back than that.
Oh yes, it's just the latest example. The Thai cave thing showed him to be a truly nasty piece of work as well.
In his defence, he was probably pretty much your average privileged ivy league (for 2 days!) nerd early on. I think it's the system of wealth concentration that is the real problem. How can you possibly maintain any sense of normalcy when everyone around you is a fawning sycophant who won't take the risk of saying boo to a goose even if you say and do something truly idiotic? Even here there's a bunch of sycophants who think he's the bee's knees, and they don't even have anything at stake.
It's amazing that any of them remain half-decent human beings really.
Must be sheer luck he has so much wealth, Tesla, Space X etc
He is a complete bell end but how do everyone elses achievements stack up?
Nothing at all to do with his parents owning a diamond mine, funding him a metric shit load of cash & him hiring some people with actual skills. Nothing at all to do with building a personal brand that makes you look successful.
As I understand it, he did get an engineering degree, plus he has wealthy parents. I think his success has really been as a visionary who set goals for the engineers working under him. Some of his ideas have been pretty silly, the hyperloop for example. That was never going to work.
Mostly, he just seems to have a toxic personality and vastly overrates his own ability. That nonsense with the Thai cave diver was just pathetic, for example. He has made some major blunders with Tesla, the quality control issues and the autopilot fiasco are down to his arrogance. With Twitter, he bought a company that he didn't understand at all at a hugely inflated price and is going to have to either keep losing billions on it each year or sell it for pennies on the dollar. I cannot see how he will turn it into a profitable business. He doesn't seem to have any expertise at all in software, yet makes crazy demands on his developers. Hard to take him seriously as an expert in AI, he's just a rich kid who likes trolling to get kicks.
Once AI figures out that we are the main problem for planet earth,I expect nothing less than some full on 'Black Mirror' moments,weaponised bees and killer toasters.
Flooded with signatures innit... "1377" 😆
Nothing at all to do with his parents owning a diamond mine,
This is quite untrue. It was an emerald mine...
Which also might be untrue.
https://www.snopes.com/news/2022/11/17/elon-musk-emerald-mine/
Hmm. People who haven't launched their own AI app yet wanting to stop others reaching mass market dominance until theirs is ready.
he just wants some time to figure out how to monetise it
Aye, my thoughts as well, if Musk is expert in anything, it's recognising that his competitors are stealing a march on him. I'd imagine that a couple of years ago he thought he was doing pretty well with AI in his cars, and more or less overnight with a couple of releases of work that other companies are ding as showed him pretty starkly that his own teams development isn't even in the same building anymore let alone the same room. I think he want's a pause so he can use the time to catch up. Simple as that.
Elon, like every other billionaire is a selfish prick, he only ever does things that benefit him.
This country (and the whole world) could do with a bit less cap-doffing to anyone rich and powerful and a bit more examination of who they are, how they got there and what they are doing.
Not many ask questions they just accept what they are presented with.
I have to admit to being a little in awe of Musks cool projects like SpaceX and HyperLoop but was very aware that he surrounds himself with smart people, that's what we should all do at work. What I wasn't aware of was the source of his initial wealth via his parents, feel a bit daft for not digging deeper into him now.
Elon Musk started life 3-0 up but thinks he scored a hat-trick.
He's clearly a very unpleasant character and only interested in benefitting one person, Elon.
Just means that people developing it with some sense of ethics and responsibility will stop, and people developing it with ill intent won’t, and we’ll be even less prepared for the results.
Very much this.
China are thought to be further along with AI than the West...
Elon Musk is as thin skinned as Donald Trump and seems to hate Bill Gates. Musk has hated Gates for ages since he found out that Gates had a huge short position on Tesla. Musk was once a huge AI supporter and promised to provide OpenAI a $1bn investment in 2018. He kept saying he would, but didn't. Gates came in later and did invest. Now, all of a sudden, AI is bad and musk is saying that Gate's doesn't understand AI.
Interesting how such an important topic that has implications for the future of human society is reduced to whether Elon Musk is a dick or not. He's just one of over 1,000 signatories.
It's an issue that should be talked about. There should be a public discourse. And talked about in a more nuanced way than - is AI going to become sentient and kill us all?
The implications are a world in which human creativity has no value. Where everyone can write a book, create software, compose music, design buildings, create art, make movies etc. Where it all becomes a mush. A drab haze.
A world where reality and fantasy are no longer distinguishable. Where there's no way of knowing whether what you see or hear is human or AI-generated. Where disinformation can be created en masse and personalised to trigger the desired responses.
Where the majority of human jobs can be better done by AI systems.
monkeyboyjc
Full Member
I don’t think Elon musk is an expert in anything……
Manipulation maybe.
I think its hilarious that humans are building the machines that might ultimately wipe them out. Nuclear weapons weren't enough obviously.
When tasked with solving climate change the logical minds will see an obvious solution I'm sure.
Good riddance I say. Humans are too stupid to have all this power. 😆
Take
10002048 rich idiots, given them a gold coin each and tell them to toss it 10 times. Probably one of them will get 10 heads.
sorry, sorry.
I asked Bard why it had given me wrong informaiton: "because I thought that's what you wanted to hear". These things learn fast.
Ignoring Elon Musk, I think there are valid conerns here.
Early deep learning models were really just doing interpolation of the input data, but there are now so large, with so many dimensions this is no longer true. Interpolation is extrapolation in higher dimensions. This means the deep learning models are coming up with new independent results and ideas. As they get larger and more complex, more properties emerge that couldn't have been predicted.
Worth a listen to this https://www.preposterousuniverse.com/podcast/2023/03/20/230-raphael-milliere-on-how-artificial-intelligence-thinks/
Musk will be trying to push down the value of a company he wants to take over. He’ll buy it with people’s money and then suddenly be an expert in AI claiming governments should invest in his company.
Where everyone can write a book, create software, compose music, design buildings, create art, make movies etc. Where it all becomes a mush. A drab haze.
We live in that world now, There's nothing stopping you or anyone else from writing anything and self publishing it. Amazon is filled to the brim with this pap already.
Musk will be trying to push down the value of a company he wants to take over.
This would make a change from his last purchase where he inflated the value, purchased it, then drove down the value.
We live in that world now, There’s nothing stopping you or anyone else from writing anything and self publishing it. Amazon is filled to the brim with this pap already.
True but I think there's going to be a vast difference in the sheer volume and quality.
There will be no divide between pappy, self-published stuff and esteemed best sellers.
AI will be able to generate infinite books in the style of a particular author or genre. And they can be crafted to the user's tastes. Removing elements they don't like. Adding subjects and themes they're more interested in.
Great works of fiction generated by anyone via a prompt. To such an extent that human creativity starts to lose its worth. A basic element of human existence that is in danger of being lost.
There will be no divide between pappy, self-published stuff and esteemed best sellers.
Well there will. Because one will be shit, pappy self-published crap written by some-one who's done a community course in creative writing who thinks that they're now the next Hilary Mantel, and well; Hilary Mantel (if she wasn't dead, obvs)
True but I think there’s going to be a vast difference in the sheer volume and quality.
I don't think that you've perhaps quite grasped how many shit novels are already published annually.
I think what you're driving at is; Can an AI write like (for instance) Hilary Mantel and pass itself off as human? I'd say that question is not wildly dissimilar to the age old question of; Is the worthiness of art intrinsically linked to the worthiness of the person (or in this case thing) that created it. Does good art automatically confer "good" status on it's creator? For instance, Are the works of Wagner still "good" even though now we know he was a virulent anti-Semite? Or Is it OK to listen to Rock 'n' Roll part 2?
Is it art if it's not been created by a human? It's not a new question. or a particularly difficult or dangerous one.
Well there will. Because one will be shit, pappy self-published crap written by some-one who’s done a community course in creative writing who thinks that they’re now the next Hilary Mantel, and well; Hilary Mantel (if she wasn’t dead, obvs)
I don't think that will happen. Somebody who aspires to write like Hilary Mantel isn't going to go on a creative writing course. What would be the point when they can generate what they want via a prompt?
Why should any young person aspire to be a writer? Or an artist or a programmer?
The ability to create is becoming worthless.
The ability to create is becoming worthless.
It already is, and has been for the entirety of human existence, for every great artist there are hundreds and thousands who never get anything published or hung in a gallery, or musicians who don't get to perform their opera or pop song outside of their bedroom, and certainly aren't paid for it. The addition of AI isn't going to radically change that. If humans want to draw, write, perform, they'll still be able to. Imposters will still try to cash in just like they have done in the past
If the public deem new works - however they're created worthy or the label "art", then it is. That's neither damaging or dangerous. That's already what happens.
