FTC investigating ChatGPT over potential consumer harm
npr.org
npr.org
Having panicked regulators with science fiction, I hope he's not surprised when they take action.
You can't poke a stick at one head and think the others won't bite you.
Altman effectively said to every level of government "OpenAI is extremely dangerous, you'd better look out for us".
All the wisdom of Bilbo Baggins reaching into the trolls pocket.
However, they shouldn't be overstated. OpenAI has had the obviously best LLMs for over 4 years now. They aren't a flash in the pan.
If he believed it was truly dangerous, I don't think he'd be working for a company that is advertising they are actually pursuing super-intelligence ?
Not saying you're wrong, but there is definitely some very noticeable dissonance in his messaging and in his actions.
You're right that there's some dissonance there, however. Especially since three of the leading labs (OpenAI, Anthropic, and DeepMind) were _all_ founded on the premise that they had to be the ones to get superintelligence right because the risks were so high.
I think he wants to scare people a little in a somewhat controlled way to onramp us to this new reality as fast as possible.
https://twitter.com/soumithchintala/status/16712671501017210... https://archive.li/rfFlW
I'm not sure the most canonical paper on mixture of experts but here's one possible:
Not op, but this is where a cheeky google got me.
Innovation! :)
It’s more likely that we will see superintelligence in our lifetime. And if the rate of progress does not slow it will be sooner rather than later.
My current estimate is parity by 2030 and superintelligence by 2035. Evidence from specialist AIs, e.g. Go, indicates that super AIs tend to occur soon after parity is reached. E.g. AlphaGo (parity) March 2016; Master (super) December 2016.
Not if we are on the wrong path, going in the wrong direction, which I think was at least partly the point of the comment to which you are responding.
There is a school of thought that something fundamental is being missed by modern AI and the (amazing) success of GPT3+ ironically risks directing us further down that wrong path at an accelerating pace.
But a collection of models, with some kind of vector databases(or something more efficient than that), being orchestrated by one or multiple master models.
A large LLM that knows pretty much everything there is to know about the world, seems like a good building block for an AGI, no matter how the rest is built.
LLM's could be like the Broca's and Wernicke's area of an AGI brain. working in unison with dozens of other parts.
> That’s possible but doesn’t look likely at this point.
Why doesn’t it seem likely?
We’ve been trying to do great things with AI for decades. We don’t seem to have an excellent grasp on why certain things work well or if the strategies we’re using can ultimately yield much better intelligence.
My impression is that we really don’t have a lot of control over progress and we could very likely hit walls and stall for many years without meaningful progress. What am I missing?
1. We are wildly underestimating the computation requirements.
2. There are theoretical roadblocks coming up such that even a very large number of smart people being paid to solve the problem won’t find a key sequence of ideas. Think Riemann Hypothesis, or Fermat’s Last Theorem, etc.
The counter-argument to (1) is that available computational resources are very high given the billions of dollars available. The one system we know to possess human-parity intelligence (the human brain) uses 12 watts and is not exactly a data center.
The counter-argument to (2) is that we’ve made faster than expected progress since the discovery of transformers, and we seem to be quite close already given the capabilities of GPT-4. Of course you don’t know that you’ve hit a roadblock until you hit it, but so far it’s been smooth sailing.
It even seems unlikely that eve after having invented/improved the LLM, that the other components are anywhere close to being developed.
If his statement were honest and not strategic, then he's utterly irrational and quite possibly self-destructive (even genocidal). One simply wouldn't continue in that line of research if one believes there was the risk of an omnicidal AGI being spawned. It's like something out of a bad Lovecraft story... if you think the spell will summon Cthulhu, the best thing to do is just stop.
>WHY YOU SHOULD FEAR MACHINE INTELLIGENCE
>Development of superhuman machine intelligence (SMI) [1] is probably the greatest threat to the continued existence of humanity. There are other threats that I think are more certain to happen (for example, an engineered virus with a long incubation period and a high mortality rate) but are unlikely to destroy every human in the universe in the way that SMI could. Also, most of these other big threats are already widely feared.
- Sam Altman, February 25, 2015
[1] https://blog.samaltman.com/machine-intelligence-part-1
[2] https://blog.samaltman.com/machine-intelligence-part-2
[3] https://www.businessinsider.com/openai-ceo-sam-altman-says-h...
> We also have a bad habit of changing the definition of machine intelligence when a program gets really good to claim that the problem wasn’t really that hard in the first place
We've done this so much recently that I'm now seeing rewritten definitions of "real" intelligence that most humans do not meet.
Chess is the obvious example. A machine capable of playing chess at a human level was supposed to indicate the advent of genuine artificial intelligence at one time. It wasn't that playing chess well means one is intelligent, but rather it was assumed it'd entail abstract planning, strategic thought, intuition, and creativity. Of course now we have software which can crush even a world champion, but none of those adjacent capabilities emerged at all.
And so I think it's also increasingly obvious that this is the same thing with chatbots. Many of us thought those 'surrounding capabilities' were finally here, even more so with OpenAI regularly demonstrating exceptional competence on a wide array of distinct metrics, such as performance on the Bar exam. But once you use the system for a while it becomes clear that its knowledge base is absolutely and unbelievably immense, but its 'understanding' of that knowledge is literally zero. It will arbitrarily create e.g. API calls that do not exist, mix up utterly simple concepts, and fail to learn from its mistakes in any meaningful way whatsoever.
I'm sure if you've used ChatGPT for anything you've run into the utterly annoying scenario of:
- "How do I [x]?"
- "Sure! That's easy, just do [A]."
- "No, you're hallucinating."
- "Oh sorry, thanks. You're right you actually need to do [B]!"
- "No, you're still hallucinating."
- "Oh sorry, you're right. You just need to do [A].
If a human, even a stupid human, acted in this way - you'd assume they were trolling you, especially one gifted with the ability for infinite perfect and complete recall.
*: (such as when GPT-4 describes what would be output by Python code that I write and feed it, if it was run, despite GPT-4 not having any access to a Python interpreter and therefore having to simulate what one would do with my code, and despite my code not being in its training set.)
> especially one gifted with the ability for infinite perfect and complete recall.
You know that LLMs don't have this, right? There is no database they have access to containing their training data. They just have the weights that were optimized in response to seeing that training data.
------
"I'm working with a new computer language. What would be the output of:
IsTrue is not true
If IsTrue is true then print IsTrue
If IsTrue is not true then print IsNotTrue"
------
It hemmed and hawed, and accurately describe what the program would do, which is basically just repeating back the last two lines to me. But refused to tell me what the output would be. When I demanded it tell me what the value of "IsTrue" would be, so I could figure out the output, I got:
"I apologize for the confusion, but as an AI language model, I don't have access to the specific values of variables in your code or the ability to execute code directly. In the given context, the value of IsTrue is not specified, so I cannot determine its exact value. It could be either true or a value that is not true, depending on how the variable is defined or assigned in your code."
I then gave it the exact same program in C#, and it unsurprisingly gave me not only a far more meaningful description of what the code does, but also the exact output - immediately.
Because OpenAI says there isn't, and there are open source LLMs which show the same (but less profound) ability.
> expert system effectively akin to an interpreter working behind the scenes?
You're describing reasoning again. No-one hardcoded an "expert system" for Python into ChatGPT. If it has become one, it is through reading about Python (and same for C#).
Also, expert system [1] needs not be in quotes. It's an 'AI' technology dating back to the 60s. It's essentially just a fancy term for hard-coding queryable domain specific knowledge into a system. As an aside, where has OpenAI claimed any of this is false? Or, for that matter, which open source LLMs produce code that's not completely buggered?
Love this sentiment, stealing it =)
Altman peddles crypto. Keep that in mind when giving him the benefit of doubt on integrity and profit motivation.
I did an internship with ElementAI a while back. They had a full lobbying division to "come up with a legislative framework with the government for ethical AI". This is quite literally just a business strategy where you get to write the rules because lawmakers see you as "their guys" while challengers are seen as reckless.
It was already a thing in 2019 and it definitely is still a thing. Sam Altman just made the gambit that congress would see him as the honest and prudent guy. Seems like that just didn't work.
There is (IMHO) no good reason to let OpenAI argue simultaneously for "this technology is too dangerous to be replicated" and for "... but if you have a credit card then you can use it right now".
But he's not. As Scott Galloway put it, he's raising a gun to grandma's head and screaming "stop me before I shoot her!" He's saying ethically-responsible things. But simultaneously acting in contrast to his words.
Altman is doing an excellent job as a CEO. But that doesn't make him an angel. Just look at Worldcoin, his crypto project, and tell me you see the marks of a humanitarian.
Because 1) the government isn't going to do anything and 2) investors will think OMG this must be profitable $$$
Maybe he really believes that he's the only person alive who can safely usher in a new AI godhead, and he feels this should be self-evident to everybody once he explains it to them.
If you consider the leaked Google Engineer memo about open source models being a threat, regulation and LLM registration would be the strategy to construct a moat.
If Sam's Gambit succeeded, OpenAI could have potentially been granted a near absolute monopoly with the legislative reach of the US government working to imperil competitors through a Gordian knot of rules and regulations which OpenAI could have been the primary creator of, perhaps even as the head of some sort private-public 'Artificial Intelligence Accountability and Trust Division.' It really just gives one that happy feeling of bureaucrazy mixed with dystopia.
[1] - https://www.wired.com/story/openai-ceo-sam-altman-the-age-of...
I see the HN headlines like "new model approaches GPT4 in benchmarks!" but they turn out to suck when I try them. What can I use today that comes close to GPT4? Certainly not Bard.
I generally run 2 tests to gauge:
1. Write me a Python script that blah blah blah
2. Write me the lyrics to a song about blah blah blah
Only GPT4 has given me impressive results. For #2 the other models manage to rhyme (except most local models), but the rhymes on other models are really basic compared to GPT4's.
OpenAI is durable only if GPT is middlingly differentiated. If it's uncompetitive, they lose. If it's vital, they face regulation and state-sponsored competition. (I personally advocate for a Heavy Press Program [1] for training American and allied AIs.)
I don't think this was ever realistic. (I don't think Altman is that politically naive.) Instead, it was marketing. It's saying you've got something so powerful it's dangerous and should be treated carefully. While everyone's debating how to treat it carefully, they concede in public that it's powerful.
Because he sold his soul to the Devil (M$) and is now pursuing regulatory capture in order to build a "Open"AI-controlled moat where no competitors can survive.
Because it's a very serious possibility(that the singularity could end humanity), and a significant part of the people who are serious about AI are extremely worried about alignment, for good reasons.
> I hope he's not surprised when they take action.
Surprised? Do you mean relieved?
Because he doesn't believe it's Science Fiction, you are projecting.
I have a belief, as a ML researcher, that there are two types of x-risk AI/ML researchers. True believers and hype-men. People do hype the x-risk because it is catchy and gets more people talking. The whole "no news is bad news" strategy, which even worked in recent elections. I think there's two dangers to these people: 1) they eventually turn into true believers (you say something enough, you start to believe it), and 2) it distracts from the risks of the current dangers. For Sam, I think he is a true believer and I'd expect this out of any CEO who spends significant amounts of time hyping people up to gather capital by promising a future AGI.
As for myself, I'm not buying the x-risk arguments. There's a lot I have to say and a lot of nuance, but to put it briefly I'll reference something Mitchel said. She noted how it seems rather unlikely that a super-intelligence, who outperforms us in every single way is also unable to understand intention behind instructions (meaning it doesn't understand us, so a contradiction in super-intelligence). The danger is handing things over to a model without human supervision and letting it hallucinate. So the danger is thinking it is smarter than it actually is, and then trusting it. Concentrating on ineffable abilities of super-intelligences distracts us from this danger, which already exists.
I’m half joking, but in the statistical distribution of good, bad and ugly things automation can do, we will need to draw some line as to what is legally actionable. For example, if I Google search recipes for explosives, Google isn’t liable for surfacing accurate information. And imo OpenAI isn’t liable if their chatbot gives the same accurate info in response to the same query in the form of a chat message rather than a link.
In his selfish efforts to build the ultimate moat around his company, Altman's attempt to get government regulation to stifle future competition might backfire.
Whilst I'd normally say it would be satisfying to see Sam fail due to his selfishness unfortunately this laser focus from the gov might bring the rest of the industry down with him.
People aren't stupid. Humans generally want to live and hope their children flourish and not be destroyed by some type of horrific event. Even...c..cc.communists.
If enough top researchers agree of the dangers, and switch from advancement to mitigation, international bodies co-operate. Things would probably be ok.
Sounds like you're placing all your hope in apathy, good luck with that.
Are nuclear weapons not "harmful or a risk to humanity"?
Everyone keeps bringing up Russia and China...they like nuclear weapons because they give them power. AI can take it away, especially if a super intelligent AI was to come into being.
Have you ever noticed how dictators like to be in control? If they actually are aware there is little chance they'd be able to control these things, they're much less likely to want them to exist.
It's just not the same type of problem. The only thing we have in common here is they can both be dangerous and cause a lot of harm, but so can fire.
This same line of reasoning is why Hinton thinks we might be able to slow things down with an international agreement, because no one wants to see things spiral into chaos. No one wins that, maybe not even the AIs.
I think the previous comment's point was this.
You said "if AI does indeed turn out to be harmful...[you] don't think other countries will be...eager to deploy it" [1]. The comment responding to you said there is no precedent for that. Before you get super-intelligent humanity-destroying AI you may get merely-intelligent enemy-destroying AI. If the second evolves into the first, and we don't know when, everyone is incentivized to develop the second out of military necessity.
My point is, everyone sits on the internet and screams but Russia, but China, but no one really knows what they think about the situation because no serious discussions have begun.
I did see an interview with Geoffrey Hinton who says that he can see Chinese researchers are concerned, he reads it in mailing lists. In my opinion that is a good start because again, I don't think the CCP is stupid and they also have a lot of influence in the world.
My point is, trying to do nothing and maintaining the status quo won't end well either.
I think most people have just given up, climate change now this, it's too much for people.
That’s the point. For a system such as this, it cannot. That’s just reality.
It gives advantage to have it, particularly when others don’t. Its development happens in silence, making enforcement uncertain. And you don’t want to get into a shooting match without it when your enemy does.
There are options between giving up and daydreaming.
Consider: a public registry of who is working on what with which data. (Good opportunity to address the copyright question.) Working group to advise, without pomp, if there is a threat and what it might look like.
We don't have an intelligent discussion because everyone speaking has a conflict of interest or no idea what they're talking about. Maybe making public more data, hard data, will help.
"Nothing can be done, oh well..."
I guess this is when humans do actually lose their humanity and go extinct, if we just give up in the face of challenges, we're done.
Also, when you disable "Chat history & training", plugins are not available to you as a paid user.
I believe Sam Altman doesn't care about his users' privacy at all. I lost trust in him completely, when he first mentioned that he doesn't understand why decentralized Finance might be useful to people. Some months later, when Silicon Valley bank was in the process of shutting down, he started crying. At the same time he supports Worldcoin.
You can't put up a sign on a rollercoaster that says it's going to kill customers if they ride and then get away scot-free when someone dies. Though it took a court to say that, I'm sure.
Same thing here. openai is saying things, and then a court is going to have to decide if that's acceptable.
Some more discussion over here: https://news.ycombinator.com/item?id=36717420
Havent heard this complaint before. And I wouldn’t really call this harmful to consumers. If anything its harmful to the subject, but consumers in general. But even so it doesnt meet the definition of libel. How harmful is it really?
Of all the complaints against OpenAI, absolutely none of them seem to be what I think really matters. And that is OpenAI tweaking it to be politically correct and geared to wherever they fall on the overton window. Its a tool. It should be sharp. And we should expect more than of its users leat we want dumber ones.
You seem to assume it's negative information about people. If it decides to say that some quack hocking raw almonds as cancer cures is FDA approved it's extremely harmful.
Hopefully it's not that bad, but there's clearly a large inbetween here and I wouldn't be surprised if it's over the line.
That's before you even get into defamation cases.
Excuse the link, the big papers have awful popovers, this one seemed acceptable...
https://decrypt.co/125712/chatgpt-wrongly-accuses-law-profes...
I don't trust AI and probably never will. I consider ChatGPT to be merely a PC version of 4chan's infamous /b/ board, which carries the disclaimer "The stories and information posted here are artistic works of fiction and falsehood. Only a fool would take anything posted here as fact."
We tried this in the 1930s in Europe and it didn’t seem to go very well. See also it’s lasting legacy in the current Italian prime minister, Marine Le Pen’s popularity in France, and AfD’s electoral successes in Germany.
NSDAP were in power for over a decade before the first concentration camp was ever built. Their rhetoric in the 30s was not as direct and violent as their later actions, that doesn't mean they weren't still Nazis.
The first concentration camp, Nohra, opened in 1933, the year the Nazis took power. The more well known Dachau was also opened that year.
Dachau was a concentration camp located in a Munich suburb that a large number of prisoners survived (but a lot did not) and were even sometimes (but not often) released from. Really bad, but nothing that awful governments hadn’t done before - in fact, was being done by our eventual allies to the East at the same time.
Auschwitz was an extermination camp (Vernichtungslager), located in a distant corner of Poland that only a tiny fraction of people sent there survived. These camps and system were a new horror for the world, and the center of the Nuremberg war crimes trials. Those were planned at the beginning of 1941 (Wannsee Conference), so yes, put into operation about a decade after the Nazis came to power.
Some places that were initially set up as work (to death) camps were later turned into outright extermination camps.
The Beer Hall Putsch of 1923 was about as effective as the January 16 assault, and just as comically incompetent. But there was a lot of sympathy for the far-right terrorists among the German judiciary, just as there is today in the US. The prosecutors and judges who could have stopped the movement in its tracks instead chose to take it easy on the perpetrators, just like today.
Now we get to wait ten years to see where, or if, these two convergent timelines diverge.
I agree, a bureaucracy banning books and reference material that ran counter to the approved narrative was quite disastrous. Hopefully we don’t repeat that mistake now that everyone is neatly siloed into digital echo chambers.
Personally, I think we did pretty well against the (specifically online) deluge of misinformation and fringe ideas for about a decade or so prior to ~2010. Unfortunately, the “problem” of internet free speech seems to have been solved and now we’re just waiting to see whether the censors or the reactionaries win and what type of fresh authoritarian hell awaits on the other side.
If you think these are the two camps, I recommend googling the words “Republican” and “book bans.”
I'd argue neither side has a stellar record today, but it's the other that's more speech stifling...
So... is this just a case of user error? Or maybe the FTC needs to require a larger font size or just better disclosure? Mostly joking, but not entirely since I do think it's too small. Here's a proposal: https://i.imgur.com/IGON20r.png
We all already treat the internet at large as a source of "usually correct but not always" information, and people just need to take the ChatGPT disclaimer to heart and not treat it like some oracle. I imagine articles like these will fade over time as society finds lies coming from the "makes-up-lies-sometimes-machine" (ChatGPT) as boring and as common place as the many websites full of lies we already have.
My significant other, an otherwise very intelligent women, will ask ChatGPT health questions. She knows it might be wrong, but does it anyways to debug her health. I try to point out that even getting suggested a bad diagnosis is very dangerous. The advice it gives has way less nuanced than say the healthMD which has it owns flaws. And unlike coding questions, you can't assume health advice is right until you prove it wrong.
Luckily, ChatGPT and Isabel almost always agree in my tests.