Hacker Newsnew | past | comments | ask | show | jobs | submit | openquery's commentslogin

I don't understand the point of this piece. There is nothing breaking or remotely interesting which we didn't already know.

Mind you I don't particularly care about Aschenbrenner or his fund but it feels to me like typical journalists reporting on hindsight without any sort of skin in the game. "Hhm, how could you be so dumb so as to trust your money to a 22 year old".

Peter Thiel did the same with Zuck and like him or hate him he did pretty well.

If you want to show how smart you are hash your predictions, post them publicly with a commitment to reveal by a certain date and then show the world how this was so obvious in hindsight.


> don't understand the point of this piece

>> What, then, to make of Situational Awareness’s plunge? ...Mr Aschenbrenner was hardly alone in such bets; over 80% of fund managers responding to Bank of America’s latest monthly survey named “long global semiconductors” as the most crowded trade. The big worry is that this trade’s whiplash-inducing reversal will have set other investment firms teetering, too.

If Citadel hadn't stepped in, Situational Awareness may have had to fire sell tens of billions of dollars of assets. Assets others have leveraged positions in. That, in turn, could have triggered margin calls and a potential credit or even bank crisis.


Is this firm big enough to cause a financial crisis on its own? Has the leverage situation gotten so dire? Is this just postponing the inevitable at this point?


> Is this firm big enough to cause a financial crisis on its own?

I'm still trying to figure out what their gross and net positions were. But taking reported figures at face value, yes, $45bn is more than enough to start a credit crisis, particularly if everyone is crowded into a small set of leveraged positions that begin fire selling.


I just don't see the comparison one was a college aged tech founder making a product for college students the other was an AI maximalist who keeps riding red without taking anything off running around the crowd cheering him on and pulling cash out their pockets pointing to the roulette wheel and saying it's always red!

He's up, but he's wiped out because he was 4x leveraged, why was he 4x leveraged? Because he's a dumb kid who doesn't understand that markets have down days or how margin calls work. The same failure mode doesn't happen with facebook or rather, when giving money to a hedgefund in a bull market you expect at least some returns or that they will actually hedge. Not fuel a mid 20s gambling addiction. As of now most of the cash is in a private position with the worst of the AI companies.

If you want to show how smart you are don't waste your time predicting the future, since it's a fools errand. The belief that intelligence makes you better at prediction ironically makes you stupid.


Mostly I found it odd to open with a comparison of hobbit to human ages when it's well known that hobbits can live to 130.


> I don't understand the point of this piece. There is nothing breaking or remotely interesting which we didn't already know.

The piece is published in a weekly magazine. It may include some interesting things that nobody knew one week ago.


"I don't understand the point of this piece."

Its purpose is entertainment

It highlights the age of the Situational Awareness founder and what he was doing before he started the fund

"There is nothing breaking or remotely interesting which we didn't already know."

It's a weekly column not "breaking news"

The column author shares his perspective on current events and usually reflects on the past. It's unlikely that readers "already know" which current events the columnist will select, what he will write about them, including which comparative history he might reference. As such, it does provide something readers "don't already know", namely, the columnist's views on the current event he has selected. As for "remotely interesting", that's an opinion. Surely this column has a readership that finds the author's writing "remotely interesting"

The Economist does at times try, incorrectly, to predict the future

But that's not what is going on in this column. It's looking backward, not forward

If the parent comment is coming from a 20-something recipient of Silicon Valley VC funding and hoping to profit from an "AI" startup, then perhaps a bias disclosure is in order

If, hypothetically, no predictions were ever shared with the public over the internet, what would that do to Silicon Valley. What would happen to the volume of HN submimssions/comments


It's like asking what is the point of Matt Levine's column as it contains no breaking news, nor anything readers don't already know nor anything remotely interesting


What would happen to "AI" startups if there was no "AI" hype, including endless predictions about the future

Of course there is no way to know unless and until it occurs


*submissions


Finally yes yes yes.

I've wanted the option to select your cookie preferences once and forget in a brower for ever.

I assume the reason this wasn't done initially was corporate pressure (most people would opt-out of everything by default).

1.2 billion exposed users × 8.17 years×365×3 banners/day×4 seconds÷36 is roughly 10-15 billion human hours lost to dealing with damn cookies since GDPR took effect on May 2018.

That's about 17,000 human lives.


Also the last generation to not frequent space. See the night sky up close.


IMO Demis has the most reasonable takes.


> More likely (without careful vetting by the folks aggregating these models) is that the quality will go down as more and more AI-generated output gets subsumed into these models.

This assumes that there aren't algorithmic breakthroughs which reduce training/inference costs by several OOMs.

How much do these models need to do before people throw their hands in the air and say, ok this is happening. The Erdos unit distance problem, which as far as I understand was approached by multiple competent mathematicians was solved by a frontier model. Sure people argue there was no novelty there (I cannot comment as a non-mathematician) but it feels like they can draw lines laterally from deep knowledge in different fields (in this case combinatorics and algebraic number theory I believe) and solve problems.

Now if you have millions of instances running in parallel, all "probabilistic", working on frontier AI research I really don't see the blocker (and believe me I wish I did).


> How much do these models need to do before people throw their hands in the air and say, ok this is happening

What is "this"? Most people arguing against some of the more fervent predictions and promises of "inevitability" are people who are using these models in day to day - they see what the models can do, and what they struggle at.

> Now if you have millions of instances running in parallel, all "probabilistic", working on frontier AI research I really don't see the blocker (and believe me I wish I did).

My genuine prediction is that you'll get a lot of early results simply because you're applying attention to some low hanging fruit of problems, but then it will drop off due to the cost of tokens and the low rate of return. This doesn't mean that the models are especially capable of novel thought, just that we haven't algorithmically brute forced a problem with known solutions.

We would be seeing more success cases if the promises were true, setting aside AGI, human replacement, etc. We would see more, better products with more features that people would use. We wouldn't be having any arguments. The human replacement presupposes the models work in ways that they don't, and until proven otherwise, can't. I've watched those who embrace it fully flounder around on projects, some have lost their mind from the constant LLM validation, and I've seen companies go all in and then pull back based on both cost and efficacy over the last year.

I'm still waiting for the success case examples applied on a scale that would make any of the predictions come true.


You're assuming that frontier AI research can "kick off the exponential" - that there exists such an exponential to kick off, and that frontier AI research can reach it soon, rather than in a couple centuries. Those are massive assumptions that have exactly zero empirical data behind them.


> How much do these models need to do before people throw their hands in the air and say, ok this is happening.

It's quite interesting to me that your epistemology is basically "God in a box is inevitable" and not skepticism.

> Now if you have millions of instances running in parallel, all "probabilistic", working on frontier AI research I really don't see the blocker (and believe me I wish I did).

The blocker is lots of things, but for one? Compute. Millions of instances running parallel, all "probabilistic" testing different theories would require million of GPU clusters.

Two? For subjects like Physics - Atoms. You need to be able to run experiments. You need access to the real world.

Then there's the S curve. Algorithmic breakthroughs operate on orders of magnitude. Eventually we'll run out.

etc. et al.


> This assumes that there aren't algorithmic breakthroughs which reduce training/inference costs by several OOMs.

Yes, if one must assume something it is generally fair to assume that things will continue as they are. Research breakthroughs do happen, but they are not something for which you can predict the timing.


> This assumes that there aren't algorithmic breakthroughs which reduce training/inference costs by several OOMs.

Of course it does.

You cannot operate based on the assumption that you will hit a breakthrough in any given span of time. Breakthroughs are, by their very nature, unpredictable—both on when you will reach them, and on whether they exist to reach at all.

It's also assuming that there isn't a hardware breakthrough that will put us another several OOMs ahead on compute.

It's also assuming that there isn't a military "breakthrough" that leaves all our supply lines either bombed or hostile to us.

When you're forecasting, you can only operate based on what is known and knowable now.

Breakthroughs are unknowable. You cannot rely on them.


This is all noise. The leaders of these companies are flip-flopping to whatever sounds best for their current agenda - hiring, fundraising, pre-IPO, etc.

The only thing that matters is if LLMs with sufficient scaling can become frontier AI researchers kicking off the exponential. Everything else is transient noise.


> The only thing that matters is if LLMs with sufficient scaling can become frontier AI researchers kicking off the exponential.

I think we know the answer to that already - LLMs show no sign of improving intelligence and instead providers are going down the ‘agentic’ rabbit hole.

There are too many things missing, like a world model, understanding, and taste (in the sense of knowing what is good and what is not good).


> LLMs show no sign of improving intelligence and instead providers are going down the ‘agentic’ rabbit hole.

I'm not sure where you're getting this. I don't work at Anthropic but Fable (Mythos) seems demonstrably smarter than Opus for pretty much any definition of smarter and they claim that Opus was used heavily in Mythos development (yeah I know take this with a massive pinch of salt).

Either way if the models are indeed helping development, even on the engineering, you can iterate on models faster and even if they're not contributing to core research yet you still have a baby exponential by improving the engineering.


Anthropic's claims about their own products have almost zero value as evidence. They have lied to our faces about stuff before, and will again if it drives the hype cycle which delivers them money.


As long as LLMs don't understand the different between regurgitating facts and making up stories, they're going to necessarily be limited.


They are taught the difference through reinforcement learning with verifiable rewards. Pretending you've solved the task or making up a story about how you solved it won't do well in that training step.


> The only thing that matters is if LLMs with sufficient scaling can become frontier AI researchers kicking off the exponential

I agree with your sentiment (about the noise), however I think this over simplifies it a bit. We may get AI that is super-human at frontier research and dramatically accelerates the pace, and still have to wait decades before it disrupts the job market (or maybe never displaces all work).

For one, the answer may depend on material science and chip manufacturing that can take a very long to build out a supply chain for even with super AI help.

And we may just find that the human mind is way more capable than we thought and even with accelerating research it's just a harder problem than anyone expected, even algorithmically.

I expect it to be a bit of both, and from ~2015 - 2025 I was in the "AI is coming for all our jobs" camp. My perspective changed last year after doing a deep dive into latest science on the human brain. (I've kept a very close eye on AI dev progress for 12+ years.


> I agree with your sentiment (about the noise), however I think this over simplifies it a bit. We may get AI that is super-human at frontier research and dramatically accelerates the pace, and still have to wait decades before it disrupts the job market (or maybe never displaces all work).

I don't see why that's the case when you have super-human researchers on tap. There are indeed physical (supply chain-y) issues to deal with but isn't the whole point that: 1. Super-human at AI research + scaling to millions of instances will probably result in super-intelligence in everything which is not AI research. (a subset of which is white-collar work) 2. Use that super-intelligence to solve any supply-chain issues you might be facing.

> And we may just find that the human mind is way more capable than we thought and even with accelerating research it's just a harder problem than anyone expected, even algorithmically.

I hope so but whenever I do, I feel like I'm coping hard and not dealing with the facts.

I'm not saying we're there yet - I'm saying the trend lines are clear.


> Use that super-intelligence to solve any supply-chain issues you might be facing.

I think this is where a lot of people's thinking goes awry. Unlimited intelligence doesn't mean unlimited resources or instantaneous implementation.


> Unlimited intelligence doesn't mean unlimited resources or instantaneous implementation.

Of course you're right. At the end of the day you need to deal with the bedrock which is the laws of physics. I could be wrong but I struggle to believe we are close to the edge of what is possible in getting the most out of our limited resources or time.

Without atomic physics, uranium would just be another shiny rock in the ground. Sand is just what covers beaches. With enough time and intelligence we've made the shiny rock power cities and persuaded the sand to solve long-standing mathematical conjectures.


> I hope so but whenever I do, I feel like I'm coping hard and not dealing with the facts.

Maybe I'm being naive, but as someone who spent 2015-2025 fully bought into the idea AI would replace all human labor within ~20 years, my updated take that it won't has a different flavor than coping.

A bit contributor to my perspective was studying the physics of the brain, and a deeper understanding of the very hard economic constraints in the current LLM supply chain.

Re the brain, we compute information in a fundamental different way that is over 1,000,000 more energy efficient than an LLM. We are also around 1,000,000 times more sample efficient at learning.

There is a lot of details in terms of specifics between complexity in human brain architecture (variety of synapses, diversity of neuron types, etc) that have no material equivalent yet.

And on the other side, understanding how hard it has been to build out euv lithography infra, and to completely retool manufacturing for a new paradigm of computing will require so much coordination, a super AI could have all the answers but we also have to deal with getting a bunch of stubborn people to align.

Some people are just coping yes, but if you really zoom into the details on both sides of the equations the gap between where we are and what it would take to dramatically surpass human intelligence is vast.

At the same time I think people overestimate how hard some kinds of knowledge work is (in absolute terms, since it's new and we haven't evolved into it as much), and how economically impactful human augmentation with weaker narrower AI can still be. (Lots of automation and prosperity but without such extreme disruption)

I think in the very long term all bets are off, but we will have plenty of time to evolve our economic systems before we get there.


“I don't see why that's the case when you have super-human researchers on tap. ”

Hahahaa this is what AI psychosis looks like


It's important not to miss the fact that AI productivity was a useful excuse for companies looking to conduct layoffs. Did some companies buy the hype? Sure - but the biggest companies would have wanted that sweet stock price layoff bump anyways and AI was a readily available justification to get it.


> The only thing that matters is if LLMs with sufficient scaling can become frontier AI researchers kicking off the exponential.

What if the answer is flatly: no? All that other stuff starts to matter a lot then.

Predicating your business decisions on a potential breakthrough that may never come is frankly insane. Imagine if at the dawn of the car industry Ford decided that it's actually a race to build the first flying car and nothing else matters.


Sure, but let’s not pretend that people treated the statements of these ceos as strategic messaging. People very clearly treated what Altman, Zuck, Amodei etc have been saying as predictions, and it hasn’t been until they’ve been proven wrong that people have started with the counter-narrative.


> LLMs with sufficient scaling can become frontier AI researchers kicking off the exponential

The cost is already outrunning the benefit to a massive amount, and the predicted expotential is not here yet. I predict it'll always be around the corner, a $1T model won't get there, but it will "look promising", but we'll sadly run out of money for the $10T or $100T model..


The only thing the $1T model needs to do is find some algorithmic speedup which allows it to be trained at $100B. I'm not saying that's easy or that it will happen but I just don't see why not.


I don't understand the logic here - a methodology to increase efficiency by 90% in training doesn't exist until it does - could you explain what you mean by "I don't see why it won't exist"? Are you seeing consistent gains by some process?


We have two worlds:

1. Cutting edge LLMs developing ASI/AGI. 2. AIs doing general knowledge work

The second world will be achieved far before the first world is achieved. And as the first path gets develolped, the second path becomes cheaper and cheaper to run inference on along with being democratized which reduces the margins for the cutting edge companies. It seems like a mad dash to go as far as possible until 90% of general work can be automated with more cheaply available tech


The leaders of these companies have no stake in it. They are beyond worldly concerns.

Napoleon got sent to Elba. Hitler ate a bullet. Your average tech CEO will still have more money than they can spend.



Getting a company to IPO is not about getting lucky once - it's making hundreds of good decisions daily. From hiring to product direction to fundraising.

Like or hate the guy you gotta give credit where it's due. (Same for Musk, etc.)


The title made me think there was a cargo subcommand `cargo copy`.


Me too, I immediately tried to look it up!


For 99% of people I don't see the usecase (except for privacy but that ship sailed a decade ago for the aforementioned 99%). If the argument is inference offline - the modern computing experience is basically all done through the browser anyway so I don't buy it.

GPUs for video games where you need low latency makes sense. Nvidia GeForce Now works but not for any serious gaming. But when it comes to LLMs at least, the 100ms latency between you and the Gemini API or whichever provider you use is negligible compared to the inference time.

What am I missing?


I'm sure giants like Microsoft would like to add more AI capabilities, and I'm also sure they would like to avoid running them on their own servers.

Another thing is that I wouldn’t expect LLMs to be free forever. One day, CEOs will decide that everyone has become accustomed to them - and that will be the first day of a subscription-based model and the last day of AI companies reporting financial losses.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: