> If it works for you, great! Just keep going. I only have the nagging question that if it’s so easy, why would the clients not do it by themselves?
Because it's not (yet) that easy, especially if we're talking a complex and genuinely useful app. Agentic coding is fast, but it's not magic.
> I’m a critical user, not a categorical denier. But there are certain categories where I see the current AI face a wall: fitting new ideas, or existing code into an existing architecture particularly.
Exactly. Which is why Joe Average still will have little to no luck vibe coding anything serious or novel. You still (IME) need a lot of active guidance, still need to push away from dead ends and propose alternative algorithms, and it takes hundreds of prompts to go from concept to what I would consider beta. (But, this is just my own experience, and it's possible I'm doing it wrong?)
I don't think that really matters. I have already seen companies generating their own code and just QA testing it, with no software engineers in sight. Or they ask us to "review" their 10k loc PR in 2 hours.
The spirit of the parent comment is true. People feel like it's productive and even if they make something worse, they are going to use it.
> At what point do we stop engaging with Anthropic’s leadership in good faith
About two years ago?
I would also note that Dario's post appears to be LLM written. Maybe... maybe... he's read so much Claudeish that it's all he can speak now himself. But I wonder if he's becoming a bit of a meat proxy.
(It's funny, I thought "pace the frontier" was going to mean something similar to "patrolling the frontier". But no, it's pace as in speed of change - everyone must slow down, right now (unless it's Anthropic, but you know we're the good guys in this, right? We're going to get someone to audit our desks!) "Pacing the frontier" feels so LLM.)
ok, he doesn't actually say this. But he is getting the desks audited...
Dario has been writing long-winded essays like this since long before AI existed. I strongly doubt he's the type of person to delegate it to AI now.
"Pacing the frontier" sounds like something that was carefully wordsmithed in committee, to avoid terms like "slow-down" or "pause" that sound much more negative. They probably did ask an LLM to generate a list of options, because why wouldn't you? But the rest of the writing is almost certainly just Dario.
If Dario is using an unreleased internal model to write, then Pangram wouldn't be able to flag it I suspect. Pangram needs enough public info about the patterns in generated text.
But HN simply ranks posts by upvotes. I know you don't intend it this way, but you're effectively suggesting that HN should not pay attention to what its userbase are upvoting, but instead curate posts by your personal opinion instead. That's ... probably not going to be a workable solution for anyone other than you, I'm sorry.
There are always a lot of HN top stories that I'm not interested in, but I just scroll past them. If I really cared, I'd use a community site that provides a filter, an app that provides a filter, or just vibecode one for myself.
Exactly. HN hasn't really changed, the industry has.
The community does a pretty good job of following the guidelines and that's helped to suppress the echo chamber effect that any personalization would add to.
AI is novel and exciting and terrifying (my how we live in interesting times!) and all of that will fade with time.
But, like you said, if somebody really wants to change their view of HN, the tools are out there.
And to prove this point: this post went from page 1 to page 4 in a matter of hour, despite significant voteship. Posts are not ordered per count only, there's an algorithm and manually applied weighs involved.
Yes, but the really weird thing is that they seem to:
a) believe that what they're creating is a basilisk, and
b) keep trying harder to do this while staring right at it
I think they're very deluded about (a) -- but if they do actually believe this (and it really seems like a decent proportion of Anthropic truly does), then why keep doing (b)?
That seems to be why this individual resigned, but I'm surprised it's not all of them. The cakeism is strong in that company.
In short, this is the believe that a god-like AI could punish them retroactively, for not having done all that was in their power to create this AI.
(A bit similar to some religious believe that a god could punish you after your death if you did not spend your live "pleasing" said god during your life)
With Roko's Basilisk, if you believe in it, the most rational thing to do is to put forth every effort to bring it into being. Because if you don't, then you will be one of its targets when it does, inevitably, come into being.
(I am not a basilisk believer. I think this is all absolute horseshit. But to understand someone's motivations, one must think like them.)
>Because if you don't, then you will be one of its targets when it does, inevitably, come into being.
Only if you don't spend 5 minutes thinking about the Self as a concept. Resurrecting someone by perfect simulation is a cool plot device for a novel like Altered Carbon (which ironically is really about wealth inequality) but not a serious hypothetical.
Even if we ignore that, no, super intelligent MechaGrok17 cannot make a perfect copy of you posthumously from your embarrassing Character.AI chats and IoT toilet data, the copy it still wouldn't be YOU. It would (supposedly) think and behave like you, but for all purposes it would be the Basilisk torturing a future stranger, or your descendants. I'm very much a fan of people not being potentiality tortured, so the rational action is to stop the creation Basilisk, or even just delay it so you aren't around for it. There's a bunch of other big assumptions that makes Roko's Basilisk really stupid, but they are unimportant. It's just theory crafting, like talking about the lore of video games or marvel comics. It doesn't help you understand their real motivations.
Roko's Basilisk is just recontextualized protestant Hell in a sci-fi coat of paint. It's the childhood anxieties of American nerds about hell and sin. The problem is that they are still culturally protestant. God doesn't exist or he's hands off, but in their minds there's still a open position for the wrathful and proactive God. If you think that you actually can create God for realsies, your very rational™ prediction of its behavior will coincidentally line up with your childhood fears.
Why is the spooky AI mad at you? Because you didn't worship it correctly.
Why do you obey it? Because if I don't it will torture me for eternity.
How will it do that? God is omnipotent, don't think about it.
Why don't try to stop it's creation? “I am the Alpha and the Omega,” says the Lord God, “who is, and who was, and who is to come, the Almighty.”
All the scifi scaffolding and stuff like acausal trade is paint over Timmy's first panic attack on Sunday school 30 years ago.
There are a couple of different formulations of the Basilisk I've heard: I agree with you that the version that says the AI will create a simulation of you to torture for eternity makes no sense as an actual incentive for you to help bring it into being.
However, the version that says AGI—as in, the Basilisk—will be created within our lifetimes proposes that it will punish the actual person.
All that said, I will reiterate that I think the whole thing is utter nonsense.
It will never not be funny that a community of self-described "rationalists" scared themselves into believing in their own version of a Slenderman creepypasta.
Not in my family or extended family. But you can see examples of this all the time when eating out. So many young couples sit at a table at a restaurant, get out their phones and scroll away, occasionally showing their partner something they enjoyed. It's rare to see them talk more than a few words on what is presumably a date night.
Companies don't really care about rules per se. They care about their risk appetite, and whether the consequences of their actions create risk above that appetite level.
The simple calculus is: will it cost me more to comply, or more to ignore? If a company chooses the more costly option, they are betraying their shareholders and being managed poorly.
If you do not fine early and often, you are almost mandating that companies ignore the legislation. If this continues for several years, then you almost make it impossible to turn around later and fine, as businesses will want to know what's changed now. You will create an impression of uncertainty and instability, which is the last thing businesses want, and the last thing you want to do to encourage investment.
If you just set and enforce the rules from the start, then everyone can relax, do the right thing and be happy about it. It's funny how rarely this actually seems to happen.
> will it cost me more to comply, or more to ignore?
Is an (unfortunately) American view of things. The vast majority of European businesses I’ve worked with attempt to comply with the _spirit_ of the laws, while my experience with most American companies is they try to skirt the letter of the law to do whatever the hell they want.
I’m not saying one is better, or that all companies in both places are like that. But it rings true in my experience.
Also Americans try to comply with the _letter_ of European laws which leads to some absurdities like the recent HN threads of people claiming it's impossible to sell goods to Europe because you have to register your packaging in every European country.
To be fair, that's actually basically true. Saying this as a swiss who stopped selling to Europe.
You need to register in several countries for packaging, there is no central thing and it's kinda obscure (in Germany there are several different companies you can do that with, from my understanding). There is a new law since August who tries to centralise and make it easier for sellers. But right now it isn't, there are several country specific small hoops and Ive experienced returned items with complain letters and threaten fees as well.
I mean, I agree. But it also has some drawbacks. Uber is a perfect example of this - they completed ignored the existing taxi and private hire laws, ate the penalties left right and centre and while they still suck they dragged taxis kicking and screaming into the 21st century in the UK. The days of being overcharged by a driver who takes a notion and a new route, or finding out the card machine isn’t working at your destination are gone, and that’s because Uber forced them to stop acting like an actual Cartel.
I would have preferred the old ways if that means upholding the rule of law. American MBAs are pure evil and toxic for a society's social fabric, and they should be kept away as much as possible, even if that means putting up with some guild-like behaviour.
Don't get me wrong but it sounds like you are applying a foreign logic to the EU system without understanding how these things usually work here.
I personally happily followed GDPR rules, so did many others. Making it generally a better place. That wasn't because of fines but because we now had a framework to follow to make the world a safer place.
I think we can say Altman is worse, his products lead to actual deaths of people just like Holmes. Altman caused children to commit suicide so I think that's worse. Altman will likely be seeing the inside of a prison cell by 2030. The public wants blood and it's either Altman or Zuckerberg.
I'd argue that the novels the TFA author reads are boring to the TFA author -- until they're not. If you are reading a novel that works for you, it's often not boring from the opening line.
Example: Anna Karenina. Gripping from the opening line (if you like the type of novel that it is; and it's fine if you don't, of course).
Or alternatively, Pride and Prejudice. Again, that opener just sucks you in (if this is a book for you, that is; and again, if it's not -- totally fine).
Example of a modern page-turner: The Hunger Games. If you are bored with this novel after three pages, you shouldn't read any further, because it's absolutely, certainly, not for you.
Or, Jeebus, Harry-Freaking-Potter. There is a reason that an entire generation grew up reading those books.
Or: "The Time Traveller's Wife", "The Poisonwood Bible", "The Road", "Interview with the Vampire", "A Room with a View", "Replay", "Of Mice and Men", "Brave New World". The list of books that are gripping from the start, should they appeal to you, can be almost endless.
Basically, if you and your book align (ugh, sorry), then there generally shouldn't be any form of "volta" (really, ugh).
There are many novels that are boring to me. I've never finished those; I discard and move on to the next one, where the author knew how to write something that isn't boring to me. Someone else might love them, and many people clearly do, but they're not for me.
Why do so many people assume that everyone will love the exact same books, and that if you're bored by a book you just have to work at it some more? Life is too short for that nonsense.
Not putting any action in the first 10% of my first novel was a big mistake, the reason is a lot of people these days "Read a Sample" and if they are not hooked then they just go to the next. Yes, the book may have ended with civilizations rising and falling but it started with a couple of uni students making a startup.
> “If you are bored with this novel after three pages, you shouldn’t read any further.”
The concept of DNF (“did not finish”) is pretty familiar to people who read regularly, but three pages seems far too early to make that call. The general rule of thumb I’ve encountered is closer to 100 - 150 pages.
Some of the most memorable works of literature I’ve read are books I didn’t particularly enjoy for the first hundred pages or more. I stuck with them mostly because I deliberately give books time to develop before deciding whether they’re for me.
My favorite example is A Tale of Two Cities. I didn’t enjoy it all that much until Book Three, Chapter 12. From there, it was smooth sailing. It’s now one of my favorite stories I’ve ever read, and I still think about it often, especially that entire final book.
That doesn’t mean every book takes that long for me. There are plenty I enjoy from the very beginning, particularly stories retold from a character’s perspective, adventure stories like Watership Down, and pretty much any Sherlock Holmes novel.
I understand the idea of putting a book down after a few pages if you don’t connect with it, and ultimately everyone should read however they want. But I do think that, by doing so, you risk missing out on some genuinely great books simply because they needed more time to become what made them worth reading.
Side note: My actual criteria for dropping a book are a little different. I tend to stop when reading it gives me a kind of vertigo or that strange feeling where your throat tightens and you start to feel sick.
> The general rule of thumb I’ve encountered is closer to 100 - 150 pages.
Practically all novels I've gotten to page 50 on I've finished. The only book I ever made it anywhere close to page 150 without getting anything back was The Fellowship of the Ring, and I think I got to page 200 before giving up (because I felt that I should be a person who liked The Lord of the Rings). It was a grueling experience. I can't fathom putting in 100 pages of joyless reading on a regular basis.
btw for people into science fiction and fantasy I highly recommend Ursula Vernon (writing as T Kingfisher) as someone who has me deeply invested almost from the get go. I mention it not just because I'm a fan, but because I consider that specific aspect as one of her greatest skills.
Because it's not (yet) that easy, especially if we're talking a complex and genuinely useful app. Agentic coding is fast, but it's not magic.
> I’m a critical user, not a categorical denier. But there are certain categories where I see the current AI face a wall: fitting new ideas, or existing code into an existing architecture particularly.
Exactly. Which is why Joe Average still will have little to no luck vibe coding anything serious or novel. You still (IME) need a lot of active guidance, still need to push away from dead ends and propose alternative algorithms, and it takes hundreds of prompts to go from concept to what I would consider beta. (But, this is just my own experience, and it's possible I'm doing it wrong?)
reply