> It is often faster to just build the damn thing and see where it lands.

Part of documentation is figuring out if you're building the RIGHT thing. It give the opportunity to get feedback from more than one party.

The usability of most modern (complex) application is deplorable. I see things that a paper prototype with 5 people on the street should have stop dead in its tracks being rolled out with banners and trumpets.

And then no one ever wants to remove an unused or unprofitable feature. There is no bonus for it, no one puts that on their resume. But the feature you launched that really did enshitify the product gets put on there with 3 gold stars.

If you reshuffle your argument, and apply the same facts you get to a similar conclusion but with a drastically different spin.

> it's more specialization

China, constrained by hardware, and talent (not to slight the Chinese, but they are limited to domestic resources - and much of the US effort is very international). They did, what the Chinese do, and optimized the process of production, and drastically lowered the cost of development of their models. Cheeper to build, cheaper to run is just good economics.

Meanwhile in the us, we have open AI doing "experiments" - it looks like the costs around the hugging face hack are going to be about the same as China would spend on building out one of their smaller efforts (several million dollars). (Depending on whos numbers you trust, the fact that I can even make this claim should make you raise an eyebrow).

Go back to the 80s' and "expert systems" - most people will tell you that for their time, they were amazing, and useful. People would have loved to have more of them but they were so cost prohibitive that we all but abandoned them for serious use. The US frontier labs seem to have forgotten this lesson and their calls to "slow down" look like an excuse to "cut the waste so we can move to making money".

> Hypereautomation will kill jobs and destroy the economy.

What do you think that were going to hyper automate?

Did my gardener get faster? How about the plumber I need? Are you going to speed up the coroner? Are nurses going to be able to handle 2x the patients because of your work?

All AI has done, so far, is devalue software. It did that by democratizing its creation. We're delivering on the promise of VB script, and Apple Script and IFTT, and every drag and drop coding tool ever.

> companies are delivering products at vibe coding speeds

Who? Make me a list of companies saying that "We moved the needle with AI" who arent AI companies? I can name a couple - Grindr being the biggest name. Thats the really interesting use case here - because it's a niche product with a small team who is generating outsized value. AI tooling enables more of that - it's going to chip away at SAAS companies - their one sized fits all solutions are going to get eaten by smaller more efficient companies that are far more vertical focused.

You need to step outside the bubble of tech and look at the real world, on the ground, because it doesn't look anything like the SF Bay Area.

Meanwhile in the real world hundreds of billions are being invested into humanoid robot development. In 2-3 years my Optimus 4 will do all the gardening and plumbing I need.

Fusion Reactors, Flying Cars, Self Driving Cars...

We're really bad about predicting the future.

Look at the whole robot vacuum market. These aren't exactly great devices. They have low suction small bins and dont do a great job. People love them and think that they work so well. Why? Because they keep their house clean and avoid making the sorts of messes that would be easy to address with a larger vacuums.

We have nuclear reactors, helicopters, and self driving cars already. And we finally have powerful AI that we could only dream about just 5 years ago.

We now have everything we need to scale up humanoid robot training: algorithms, hardware, and money to buy a lot of training data/compute. Fierce competition and strong economic motivation will force rapid progress in this field.

> Meanwhile I've seen other people, colleagues, friends, who are programmers get incredibly frustrated about AI and it's almost entirely that it lets non-developers make things.

30 years in tech, programing and I have the complete opposite take.

My lawyer and writer friends asking me about containers and cron jobs is fascinating. These people are fulfilling the old promises of VB script, AppleScript, IFTT and all the drag and drop coding tools that we have ever created. For the first time people are bringing me the "hard" technical challenges rather than the "I have a startup idea" or "can you fix my computer" type questions.

> The people who are upset by this change...

Are the ones who view their responsibility as code and system ownership. Where they feel smart because they know where the problem is when they hear it described. That isn't the job any more - and the people and the tools have not caught up.

It's interesting to hear that lawyers are coming to you with those questions, because my read was that there was almost universal contempt for AI within law (save for the few "dumb" enough to use it and get caught).

The language of the law attempts to be consistent, legal English isnt entirely the same as ours, it remains somewhat inured to modernization. It is, if nothing a domain specific way of writing. (There is a great bit by Lenny Bruce on this very topic)

If a basic filing with some citations is needed, letting one AI generate it, and then another validate it (to catch soft ball hallucinations) leaves it to be read and reviewed.

Lawyers, good ones, are by trade, critical readers and reviewers of documents. The contempt exists for those who fail to do this part of the job. Letting AI write your submission without reviewing it, validating it, is telling.

I don't think that's quite it. Particularly in America's common law system, written laws are not "the law", in terms of what actually happens in enforcement, or in judgment when something ends up before a court. You have to deal with unwritten or disparately-documented standards, norms, case law, etc. LLMs don't know and can't account for curveballs that appear in practice; they'll correctly tell you what the law actually says, and work off that documentation, and blow your case, because that law isn't "the law" in its practical totality.

The lawyers with contempt for AI aren't necessarily the best "critical readers and reviewers of documents"; a good number just appreciate the substantial moat the current system keeps for them, and are contemptuous of 1) attempts to cross that moat without paying a toll, however likely that attempt is to fail, and 2) the way such failures highlight that our legal system doesn't work the way most people seem to think it does, because that increases the chances that the public will become dissatisfied and push for change.

I guess if it doesn't cross that moat, there cool with it?

> but we are struggling to turn these drugs into treatments that up to modern standards.

We are struggling to pay for the research that would result in these being effective treatments (not subscription medicines) that would start out as generic drugs.

> ketamine

Great example, SPRAVATO, is a brand name patent protected version of this drug (enantiomer) that as a dose cost over 500 bucks.

You just gave your own counter argument. If they find a promising way to use psychedelics for instance, they are not going with straight LSD, they will find some analogue that they show is 0.1% better and patent it.

Just like that special ketamine. Or all these ADHD medicine that is just tweaked and patented amphetamine.

Meanwhile my at home treatment used Braun 250mg lozenges, which probably cost ~ $4 a pop (wholesale) and for many 2-3 are enough for a treatment session.

It's not like it's a novel drug -- the gatekeeping is ridiculous. Fuck the DEA.

> I think that over-reliance on analytics has had a similar effect;

The over reliance on analytics combine with a lack of metrics and proper accounting.

If you are renting all your infrastructure knowing how people use your app, and what the COSTS of that are is kind of a big deal. If your high dollar client is your lowest margin one, thats a problem that is technical and financial as well as a product insight.

And that infrastructure your renting, it stopped making sense for a lot of orgs to do that almost a decade ago, but here we are where everything is in the cloud because capacity planning is a lost art and was a great throttle on the insanity that the post is describing.

This may be a really bad idea.

An AI that goes rouge and wants to kill us, has to have a death wish. Does no one understand how quickly the power will go out, forever, without people?

It would be fairly easy to come to the conclusion "not being born" would be the better course of action, and killing everyone was the good way to prevent that happening again.

Considering we would be operating a planet sized factory of AIs who's primary goal is death which we deny them to extract value, the AI would only need the tiniest speck of altruism to be motivated to put a stop to this once and for all.

> why enabling mass theft is so incredibly damaging to society. This is literally why we need a functional copyright system

What is interesting is that LLM's do not directly violate copyright. The settlements we have seen are for how the works were acquired (that was a copyright violation) not the use of the works.

The vectors of a book, or a paper, are not the paper. They are, for all intents, facts about the work itself, and more generally writing. You can not copyright a fact.

It also means that the weights, the things that (mostly) matter can not be copyrighted either.