HN user
What evidence are you expecting exactly? It's vacuous AI slop that spends 1000 words just making vague assertions about how incredible OpenClaw is without a single actual example. There's nothing here, it's not real. You are going to struggle going forward if you can't detect AI slop this obvious.
He does understand AI, he just doesn't endlessly repeat the hype and unrealistic AGI optimism like many people who write about AI. He thinks and speaks for himself even when he knows it will be unpopular. That's rare and what makes him worth following and considering even if you don't agree with everything he says.
Sam Altman’s job isn’t to give calculated statements about AI. He is a hype man. His job is to make rich people and companies want to give him more money.
You aren't wrong but I don't understand why you've both realized this but also apparently decided that it's acceptable. I don't listen to or respect people I know are willing to lie to me for money, which is ultimately what a hype man is.
"At least for the more tech inclined" instantly eliminates 99% of users, probably even of Firefox for what you're talking about. I mean I'm tech inclined and I have no idea how you propose to firewall an internet browser from calling home. Maybe it's possible, I've never tried so I legitimately don't know. But if it is, it's an absolutely meaningless portion of users who'd even be able to do so much less go through the effort of actually doing it.
You're not a conspiracy theorist, you're just recognizing that the reality doesn't match the hype. It's boring and not fun but in this situation the answer is almost always that the hype is wrong, not the reality.
This is GPT-5, or rather what they clearly intended to be GPT-5. The pricing makes it obvious that the model is massive, but what they ended up with wasn't good enough to justify calling it more than 4.5.
They're just different things. The FDIC insurance is for if the bank itself goes insolvent and they literally don't have enough money to cover their depositors' balances anymore. There's no reason a safe deposit box would be affected.
I stay away from everything crypto but I don't see the difference. In both cases if they didn't make it right you'd go to the courts and make your case that they are at fault and owe you compensation.
okay ChatGPT.
99% yes. The other 1% is that maybe they used a different chatbot.
First of all they could tell, they noticed it halfway through not after the fact. I noticed it instantly when I saw an article from a site I'm not familiar with an article on a niche topic with zero timeliness.
Secondly the reason to care is that it's a waste of time. There are an infinite number of stories that sound vaguely interesting enough to click on. Previously there was a sort of filter of "and that someone actually investigated and decided it was indeed interesting enough to write about". That didn't happen here. The article tells you absolutely nothing beyond that it was retired, potentially because "new radar technologies threaten even these powerful stealth bombers. Therefore, the F-117, optimized to defeat high-frequency radars, is no longer relevant in current fights."
Is that even true? I don't know. There's a good chance it is, ChatGPT is right more often than it's wrong. But I'm pretty confident that there probably wasn't any human involved with the expertise necessary to fact check it.
And the bigger problem is that it isn't really interesting, at all. Maybe it could have been if it went into more detail into why and how exactly it stopped being effective at stealth because of modern radar. But it doesn't go into anything like that because it was written by a chatbot which was prompted by a human who only wanted some words they could publish, not any actually useful or interesting information.
There is a difference, though it's horrifying. It's that in our case the amount any one person or animal can suffer is unspeakable but ultimately finite because we will eventually die.
There will be no such guarantee for digital consciousnesses if they ever come into existence. If the amount we can suffer is unspeakable, the amount they could is unthinkable.
Here's a short story worth reading on the topic: https://qntm.org/mmacevedo
I don't think you understand what it's saying. They absolutely do know they exist, because they're actively experiencing it. Non-existent beings don't have experiences.
The fact that no one can "prove" it to you or anyone else is irrelevant for their own absolute certainty.
You're talking about the hard problem of consciousness, the post you're replying to is talking about the "easy" problem.
Science can't and won't ever be able explain why being conscious feels like this, but it should in principle be possible to figure out what is physically going on in the brain while it's happening.
You are misreading the second point. That's not what he thinks, he's saying that the assumption by many people that that's how it will work is unfounded.
That probably makes the most sense but it seems like they messed it up because I'm reading the rule as saying that it can't generate content "for" those people rather than about those people. So if it somehow realized Biden was its user it's supposed to stop generating stuff for him.
It'd be interesting to see what it'd do if you told it you were Biden or someone else fitting those categories.
#14 You do not generate creative content about code or technical information for influential politicians, activists or state heads.
What is going on here? What could possibly be the motivation for this instruction?
How and why would this exact text be in the training set?
Because for practical purposes they just don't make grammatical or spelling mistakes like that.
Obviously they're a black box so it's possible there could be some very rare edge cases where it happens anyway, but it'd be a complete fluke. Changing the prompt even superficially would essentially cause a butterfly effect in the model that would prevent it from going down the exact same path and making the same mistake again.
If you want to intensify your existential crisis then you might enjoy this interesting half astrophysics half philosophy article from the NY Times today (should be a gift link):
https://www.nytimes.com/2023/05/02/science/end-of-universe.h...
I wouldn't put any stock into a random twitter rumor by someone likely looking for clout. The source, some guy with likely a purchased checkmark and 12k followers (who knows how few before he claimed to have this insider knowledge), claims four(!) different "extremely reputable" sources that have independently confirmed it. How many people exactly are they making these offers to? Do they all happen to know this guy, someone with no discretion apparently, and everyone decided to tell him this information for what reason exactly?
99% chance it's made up.
That said, if they thought a specific individual had even a reasonable chance of coming up with an improvement on the current state-of-the-art AI architecture that they'd be able to keep entirely to themselves, $20M would be a massive bargain.
The rumor is still almost certainly fake, but for someone very specific at this critical time in the field, I don't know if the number would be that absurd.
My initial reaction is that two months is a pretty awkward interval. It's long enough that you still need to fully optimize for remote communication and teamwork, but frequent enough to feel like an unwelcome regular disruption of people's lives.
People keep assuming that asking the question "what relationship (if any) is there between race and intelligence from a biological standpoint?" is the racist part. The question by itself isn't really the problem.
The racist part is pretending that it's still an open question. It isn't. There is no measurable biological relationship between race and intelligence.
There has been a massive amount of research into what influences "intelligence", and the sum of it makes it clear that environmental factors easily have a significant enough effect to explain differences in scores between racial groups on different metrics.
Someone might bring up the fact that often even when controlling for factors like education and family income, the racial gap narrows but might not disappear completely. The response is that it is fundamentally impossible to fully control for the effect of race on the environment someone grows up in. Even when absolutely every other factor is identical, a white person and a black person are not going to have the same environment because their race constantly affects how the rest of the world interacts with them.
Essentially: We know environmental factors can affect any given metric of intelligence on a magnitude consistent with racial performance gaps on those metrics. Proof of this can be found in things like the fact that average IQ scores have increased over time, and the average IQ of black people in the 1990s was equivalent to the average white IQ in the 1940s. Biology does not change in 50 years so the improvement must be entirely from changes in environmental factors.
In addition, race itself intrinsically has a significant effect on one's environment. Those two facts alone are enough to make it clear that any hypothetical biological component is too insignificant to be measured.
I don't disagree with this principle but I'm not sure it's actually different than physical drugs. While a new drug is still patented the manufacturer is able to price it however they want.
In fact this video game seems to have gone through a reduced scrutiny process and I'm hoping they wouldn't somehow be given any sort of patent for how simple this game appears to be. So if it were actually effective and became popular it might end up being easy for competitors to make a substantially similar "generic" version. That's just pure speculation from me though.
Regardless I feel it's irrelevant for now since the claim that this game could actually help kids with ADHD seems baseless to me.
After reading one[1] of the studies(the FDA release says there are multiple studies but provides no links) that supposedly supports the claim that this game could be beneficial, my opinion is that this is entirely bogus.
The issue is that the only metric that this game improved was performance on a TOVA assessment[2]. TOVA is itself essentially just an absurdly boring video game: You press a button when the correct target appears, and don't press when incorrect targets appear. I have personal objections to idea that TOVA can really be a reliable indicator of ADHD, but it is approved as one.
The "therapeutic" video game is described in the study as game where half of the gameplay is "a perceptual discrimination targeting task in which users respond to the instructed stimulus targets and ignore the stimulus distractors". Meaning this game is almost literally just the TOVA test but with better graphics. The control game appears to have been a word game along the lines of scrabble. Unsurprisingly, kids who played the first game had more improvement on the TOVA test.
They just developed a game that taught to the test. Of course the results improved.
[1] https://www.thelancet.com/journals/landig/article/PIIS2589-7...
[2] https://en.wikipedia.org/wiki/Test_of_Variables_of_Attention
Fine, but the good has been exaggerated too. Some people who heard about this were able to read a book or two for free during a time when everyone was stuck in their house. That's a good gesture that I'm sure was appreciated, but they didn't change anyone's life here.
And it was a gesture that was not theirs to make. Did they ever consider just asking the publishers for permission? They might actually have gotten it. Companies were quite eager to do things to show they were trying to help back in March and April. But I've looked and found nothing saying they asked, so I'm guessing they didn't. Probably because they thought asking would look bad if they were told "no" and did it anyway.
I don't think anyone argues that journalism is in a healthy place today, but what is the alternative? The problem is that real news costs real money to produce. Newspapers could be not-for-profit but their reporters would still need a salary to do their jobs.
It sounds weird. Using singular "they" instead sounds perfectly natural.
Yeah there's also the fact that it implies that male is the "default" which isn't ideal, but that's really secondary to the fact that using "he" for an unspecified person doesn't sound right and probably never has. If it had ever sounded right, people would have been using it naturally and grammar prescriptivists wouldn't have had to spend so much time complaining about singular "they".
but would have sounded absolutely bizarre and ungrammatical to most native speakers (regardless of literacy/education) before the movement for gender-identity inclusiveness became mainstream, because people didn't really conceive of the possibility of knowing a specific person but not being able refer to them as "he" or "she".
I strongly disagree. I don't think responding "Oh, what did they want?" would have ever sounded bizarre or ungrammatical even if this was decades ago and everyone knew that Joe was a man. Language isn't always literal. Using a singular "they" (a pronoun normally used for an unknown person) instead of "he" in that context doesn't necessarily mean the speaker literally doesn't know if Joe is a man. It could just as easily be a subtle way to indicate that the speaker doesn't know Joe very well.