kaffekaka
kaffekaka on Hacker News
Instead of just asking the researchers when RSI would be achieved? Sneeky!
Are a majority of Anthropic developers working on SOTA models and RSI? I would have guessed not.
> Tao didn't write this article.
True, but you referred to Tao yourself: "The host of the blog, Tao, was an early booster. He didn't think through the implications, and seems to have expected it to remain unable to do interesting work. Today he's on a "war footing"."
My comment was in response to this.
Product Market Fit (of AI) is highest in software development.
Absolutely. This is an aspect of the tech bro dictatorship.
Yes, this. Unless there is a true singularity where no more questions are ever asked, there will be definition be infinitely many problems to solve, and most of them we have not even glimpsed yet.
Singularity? Probably not.
Tech bro dictatorship? Unfortunately a bit more likely.
None of the above? Then we have a small shot at a great future.
Having a machine that solves hard problems must be a good thing.
It is weird. From what I can tell, you and I have quite similar expectations on what AI will be able to do in the future, but for some reason you are much more dismissive of Taos opinion.
I perfectly understand his point and it is not about being selfishly frustrated by a machine being better than him. Is that really your interpretation of the discourse?
Mathematics as a field will not cease to exist, and not everyone will be interested in (or qualified to) managing the fields progress and its applications.
> Their point may also be (re: wine), that some devs have been conditioned to think a certain token vendor (vineyard) is of superb quality, while if given a blind test (tasting) of the agents on a real world task, they would not notice the difference.
Yes, this was the core of my point.
Sorry for being unclear. I mean that we humans have a tendency to believe we can detect nuances in the world when it is in fact just a case of internal bias. I simply question whether most people can actually detect whether they relly need the best model for their task. But of course, everyone believes that they themselves have this ability even if people in general do not.
At home I have cancelled both OAI and Anthropic subscriptions since some time and now only have a Mistral subscription and credits on openrouter. I use Opencode.
At work someone else is paying so we all use Claude Code. It works great but I want to set myself up to have alternatives the day the tech oligarchs decide to tighten the screws.
Open models are perfectly fine to use, but at the moment of course much more expensive than the claude subscription if you do lots of work. The Mistral subscription at $20/mo includes API credits of $20 though.
How about just practicing treating humans like humans and machines like machines then?
It is a very human trait to believe/feel "what I do is special so general rules do not apply". The bias is very strong.
The same way people have opinions on things like for example wines without actually being able to discern them in a blind test.
We humans want to think we can have strong, well founded opinions even when we cannot.
Yes. All of this sounds wise to me. Developers must be allowed and encouraged to expand their scope of responsibility in relation to how AI expands their code typing speed.
Blindly crushing jira tickets at 4x the speed will lead to crap, but investing the gained time into being more strategical will lead to higher productivity AND for many people greater job satisfaction.
Correlated for sure. But the cause and effect is unclear.
Shipping lots of good code by necessity means you ship lots of code.
Shipping lots of bad code, too.
But if you only know that you are shipping lots of code, which of the two is it?
IOW: by increasing number of shipped LOC, you will not magically make it better.
Do you think their endgame is achievable? Honest question, your reasoning seems balanced.
I could not find this in the S-1?
Exactly, the important parts are spread out over the whole text.
But also: I am bad at skimming, better at close reading, and LLM text is unreadable for me as well.
I have on a few occasions tried, and failed, to instruct Claude to write in a more comprehensive way. So while it may be possible to affect it but prompting, it is not certain to work and to remain in effect for a longer time.
I agree, well put. Data dense but information sparse.
One is bombarded with facts and struggle to get what the text is actually saying.
I have tried getting the LLM (Opus 4.8 or some 5.x and some other model) to write more human like, and also analyze its own text. It was hilarious, it could describe the problem quite well, but was helpless to actually do anything about it. "Oops I did it again, sorry!" LOL
The problem is when the AI has to describe something you do not understand, like a complex piece of code or a series of unknown changes. Or all these horrible "sharable artifacts" that are spreading like a disease.
Why is "vi går hem till mig" incorrect?
Opposite where i am. PMs would gladly ship good enough while some dev colleagues are more strict so to say.
Yes, my point is that when people say they are more productive, and that consists of more ppt slides, there is confusion and slop happening.
A difference is the output of software development can be immediately applied in an automatic way - production leads to delivery as a matter of fact. The whole flow can actually be automated.
Producing legal documents on the other hand is only part of the process, a lawyer or similar must still apply the produced text in some human driven context.
I think the question discussed here is more: it didn't disrupt everything in ~four years, will it take another two, five or twenty?
Is "time to prepare report" something that will actually increase overall productivity? Or does it really not matter that much?