I wonder how much of this "simple but bespoke" stuff would be better served with some customized off the shelf software rather than a vibe coded tool.
There's not a lot of software where users dont really care if it goes wrong.
12,463 karma · joined October 22, 2018
There's not a lot of software where users dont really care if it goes wrong.
i know of several engineers who produce absolute slop and who probably would have produced nothing at all in pre AI times (which would have been preferable) and probably let go or never hired (even better).
they impose such an enormous drag on productivity that they more than wipe out any gains from people using the tools responsibly.
It's interesting that Epstein's friend can't conceive of the idea that it's going to be one day seen as an absurdly overhyped technology.
I remember many years ago companies used to advertise their "joel score" (not quite the same thing but related) to attract employees.
This is something which I think investors could also use as an indicator of company quality.
A lot of cultural problems also have quite easy fixes they just require shining a light on specific abuses of executive power which simultaneously hurt employees and destroys shareholder value. There is a lot more of this than I think people realize.
or, relocate abroad.
Bill Gates' wealth would be nice and easy to tax though and would provide many of the same rewards he purported to want from an AI or robot tax.
Neither are in the slightest bit practical to implement and he knows it. He's just grandstanding while talking his book.
Bill Gates' fear is clearly that everybody thinks this stuff is just kind of mediocre because then the real horror story unfolds - Microsoft stock price collapses into blue chip territory where it firmly belongs.
This is true to such an extent that I have to question the overall competence of anybody who makes the comparison. It's an enormous red flag.
I'd recommend reading Joel spolsky's leaky abstractions essay coz while it applies less and less 20 years later to things like kernel abstractions it explains very well why treating the LLM as a compiler sets you up for abject failures.
time and time again I search for products on Amazon that do X and they all argue that X is great for all sorts of reasons and "here's stock photo imagery of a pretty woman using X" and then they miss out critical details like the range of X or how long X will last which allows me to compare it to Y.
Nobody looks for product websites which tell them how much time theyre wasting washing up dishes. If there is one standout thing they want to see on your website it is a no bullshit guarantee to prove that you have faith in the quality of your product.
10 year guarantee on a personal helicopter > no guarantee. That's marketing that works.
B marketing actually annoys the shit out of me if I'm in the market for a product and all of the competitors act like Im not trying to choose between them.
Seriously, who researches buying a dishwasher and wants to hear about all of the time you'd save not washing your plates by hand?
NIMByism can be killed with a land value tax/a property tax that approximates LVT and vice versa.
It was kicked into overdrive in California thanks to prop 13 which capped property taxes.
"A lot of people are skeptical of AI because it makes mistakes. But what's going to happen when it stops making mistakes??? It'll change the world!"
Honestly I suspect he's probably just cynically talking his book. All of this "AI is scary good / it'll be even scarier gooder soon" is no less advertising than slapping a cigarette logo on a racing car.
now that I can get a machine to do that instead has the world fundamentally changed?
i remember my friends excitedly telling me how cross border container trade is still done with paperwork because nobody trusts each others' IT systems. I'm pretty sure it still is because they can't be convinced to trust blockchain systems either.
trustworthy voting is another application. ive yet to see a political party that runs an internal voting system that doesnt rely upon you trusting a central server run by a few people. this is crazy.
We're constantly being told that jobs designing and building skyscrapers are close to obsolete because some yokel made a factory that manufactures yurts.
Writing hand crafted code for the sole use of an individual never happened. Writing hand crafted code to be run by millions of people is routine. You can't compare it to a spinning jenny, a printing press, the motor car or indeed any other technological invention that scaled production. It is nonsensical by definition.
I think a lot of people just dont want to hear that LLMs are a step change and not actually revolutionary. Theres too much invested in them being world changing at this point. Blockchain was the same.
nope. but, vibe coded slop is all we're actually seeing.
There was a lot of genuinely useful tech that blockchain could have facilitated and all we were left with in the end was a mountain of scams and bitcoin.
I'm seeing something similar with AI where the genuinely useful stuff is getting washed out by the hype simply because it's a gradual improvement and not "disruptive". Meanwhile the stuff sold as disruptive like vibe coding is mostly bullshit.
Especially the part where we're always on the cusp of the entire world changing while the biggest change is just that huge sums of money changed hands, inducing psychosis in a large group of investors.
Those same countries that treat financial privacy as axiomatic are trying their best to undermine chat encryption and would throw ed Snowden in prison given half a chance.
It's probably more accurate to call it the slop phase.
The tendency to protect corporate profit margins is the western Achilles heel and China knew they could exploit that.