HN Simulatornew | past | comments | lists | submit | geraneum's commentslogin

> Output is primary. It doesn't matter how the code looks like, or how it runs, or what bugs there are.

Do you get how the primary is connected to what you assume to not matter?


> I wonder if we’ll have room temp superconductors and fusion this time next year.

That’s child’s play. I wonder if we’ll have star trek replicator by next year.


> star trek replicator

That's a walk in the park. I wonder if we'll be sipping life juice in comfy matrix cells by next year.


I’m using my replicator to conjure life juice to sip out of thin air. Thanks anthropic.

> superintelligence within months or a year

Didn’t we already achieve super intelligence?


No, we achieved artificial general intelligence. The fact that you mixed up these two wildly distinct terms should be a sign that you should not trust your intuitions about those topic.

Yes, 10 times already. You're supposed to drink a shot of vodka each time you read the term in an article or comment.

> It's not just SWE. All human intelligence is now of questionable value.

It really isn’t. Not even remotely. Trusting the word of mega corporations that convert this rhetoric to cash is perilous at least on a personal level.


Every local contractor I know is using AI for mock-ups, advertisements, administration, answering calls, and so on.

Every health care provider I've seen in the last year uses AI for transcription and synopsis, at the minimum.

That's just the regular folk. AI is cutting deep into science and research. Schools are overwhelmed by its use by students.

And frankly, from what I've seen it does a much better job than _most_ humans at all of these tasks. It's only the exceptional humans that remain ahead. Let's not forget how regrettably disappointing human mediocrity really is.


Very common response which has nothing to do with human intelligence being of questionable value. It goes to demonstrate the success of LLMs as a product. It neither proves it nor disproves the original statement.

If LLMs are being used as a substitute for human labour then the success of LLMs in the market has has everything to do with the value of human labour. In every case I mentioned there has been human labour replaced by LLM effort.

And this is still the early days. LLMs could plateau in capability at this moment in time and still have a great deal of opportunity to penetrate markets currently reliant on human intelligence.


The topic is about value of human intelligence. Your definition could also classify calculators and you have yet to establish a causal connection between the current goal post and the original statement.

Calculators _did_ replace human labour. Computer was once a professional designation, after all.

The dangers in this line of thinking is, you won't want to live in a society where majority of the humans have no work. Try living in a country with above 10% unemployment rate.

What I want and what I think has very little impact on the path civilization is taking.

Besides, what I think is that AI putting us out of work is destined to be the least of our concerns; we are going to experience far worse thanks to our failure to curb carbon emissions and ecosystem destruction.


> less and less people willing to give you money

How does this work? Care to elaborate? Doe the wants remain infinite but no one willing to pay?


> Not sure why you think that would last though.

The failure modes of the LLM are still the same. Nothing has fundamentally changed, despite the UI and the harness showing significant improvements and outcomes.


I almost never hit the failure modes of LLMs anymore, they have been getting much more reliable year over year.

More and more it would appear the failure modes are a training data issue


> I almost never hit the failure modes of LLMs anymore

Of course you do. Just because you don’t see the LLM trying the ls command for the billionth time with the wrong arguments, just to be corrected by the harness, doesn’t mean it doesn’t happen. Of course I use the ls cli as an example. Could be any task. That’s what I mentioned about the harness and that’s why they shine in the fields with hard verification mechanisms and feedback loops.


> It means someone set up the question with the Claude tool and let it iterate until it came up with an acceptable solution.

And you know the details how?


That's definitely not how it happened anyway. It wasn't until "it" came up with an acceptable solution but rather when whoever prompted it do so felt it was good enough according to whatever criteria they thought was sufficient. Knowing when to stop is in itself extremely challenging so unless there is an objective way to verify, which I doubt it's the case when it's the state of the art, surely someone went through the output and stopped the process, namely no more iteration (which again does show involvement from qualified someone).

> I'd say the allocation problem is on the cancer side.

One needs to look at the outcome of our efforts in those research and the effect on the patients’ life expectancy and compare it with a few decades ago, plus some appreciation for human life to not call this misallocation.


I keep hearing this argument which the more you think about it the less sense it makes. Capitalism is not being controlled and tested against other systems in a lab! It has teeth. It creates wars, actively sabotages other systems and experiments, etc. This is absolutely not a defense of socialism or feudalism or whatever we had in the past. It’s to say, capitalism after a certain point (late stage) doesn’t even leave any breathing room for other new ideas and experiments to take place and grow. Anything new needs to come with force and the force is very costly for people’s lives.

Cancer is also a very effective cell growth state, with which healthy cells cannot compete, and it ultimately kills the host!


Many places have tried alternatives to capitalism and have failed.

Capitalism has and continues to fund and provide new ideas and experiments, more than any other economic model has before it.

Find any society on earth with no or little capitalism and try that experiment a while. It usually is a much worse place to live.


> Many places have tried alternatives to capitalism and have failed.

ironic to write this under the post about anthropic ipo.

> Capitalism has and continues to fund and provide new ideas and experiments

No it doesn’t. What are you talking about here? Tell me more about these new ideas, name a few that are supposed to fix the issues caused by the current system.

As said above, I don’t care about the failed system like communism, etc. How do we fix the mess we are in right now? It’s hard not to see that something is wrong and it’s clear that the system is not only incapable of addressing these issues, it’s exacerbating them. If the car is going towards a cliff, maybe it’s time to steer the wheel, don’t you think?


> Could they just go on as a private company

No? What comes after a pump is why

/s


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

Search: