glimshe

"Means testing" is the bad Social Security idea of the year. SS isn't a welfare program. Trimming benefits, changing retirement date (for all) and uncapping contributions are much more palatable to voters.

You're mixing a bunch of different "people" in the same bag. I'm not the one saying "there's never been anything like this before".

The investment is massive, but not bigger than other projects by an order of magnitude. Major wars were bigger than AI. The dot com bubble was 2-3T dollars.

There are two key elements in the current AI fearmongering:

1) People think they can predict the future

2) People think humans can't adapt

The number of disruptive "game changing" technologies throughout history is immense. There's no question some people get negatively affected, but humanity as a whole has prospered because of these disruptions.

Literally every time, including during the industrial revolution, people thought it was game over for society, good manners, the youth, the planet etc.

Keep cool and try to adapt to the inevitable change. The next generation will certainly do it and leave you behind if you stay in a corner denying the new reality.

Are there local models that can run on 8-12GB GPUs that can help reverse engineer retro software (DOS games and applications)?

Should we blame this on memory companies or the AI companies bidding for memory? How should the memory companies have acted differently?

I have a question for people who understand this better than I do. I'm not a climate change denier or indifferent to it.

Q: is it possible that climate change could be a net positive for a significant number of people, even if a global net negative?

My concern is that some people could be encouraged to promote CO2 warming because of positive local effects.

What do you mean by "because they require a lot of detail"? I assumed that most classic RPGs are an engine reading tiles and some intermediate scripting/config language for the behaviors.

Firefox, I used to love you but you can't UX your way out of a decline.

This is great. But maybe part of the motivation is that 6-Sol wasn't as good as initially advertised so they needed to tweak it. I felt a clear degradation in quality in some simple refactoring tasks vs 5.6-Sol.

Never mind that troll.

Given your work described on this great article, how do you see a path where the LLM produces a game replication that a human could directly understand and manually improve - even if arguably unnecessary?

Take a game like Wasteland, which has a terrible modern port. I'd be interested in creating a modern port as a hobby while manually making improvements myself. But that would require code that is reasonably modular which a human can understand. Is this possible using your method and current models?

MongoDB can't deliver high availability.

"solved coding" is type of thing you say if you want to sound smart.

Saying that LLMs have "reduced the cost of coding" would be boring. And using your analogy, pencils, typewriters and computers have all reduced the cost of writing, but writers are still around.

At this point I'd rather use a Claude-written database system than use MongoDB at any price.

Roko's Basilisk is one of the most stupid concerns I've ever seen raised by otherwise very intelligent people.

I don't think it needs an elaborate refusal. I'll just say that in the universe of concerning future problems, it's just one of many, much, much more likely outcomes.

I'm nearly infinitely more worried about an impact event, which is a guarantee some time in the future, or global warming, an uncomfortably high likely outcome, than Roku's Basilisk or anything resembling it.

Ben Felix is also pretty good.

Fair enough, although nobody thinks Mercedes cars will possibly replace/destroy humanity.

This is the first time a corporation has spoiled a word for me.

"Copilot" is a term like any other and when I first heard the name, I thought it was an alright, if uninspired, choice.

Now I can't hear the word "Copilot" used in any context without having negative feelings due to Microsoft's saturation campaign.

Speaking of this, I also wonder about present and future people named "Claude", a mainstream name in some parts of the world.

It's not that clear and still involves coding. But as I start coding and modules/APIs/classes take shape, it gets to a point where I say "ok, from here on it's just filling in the blanks".

But filling in the blanks can still involve a lot of grindy, time consuming work where you iterate through collections, build parameters for other APIs, check invariants etc. All of that stuff is boring to me.

What I'm trying to say is that most coding tasks require "senior level thinking" and "intern level thinking". The latter can sometimes be the majority of the work and I don't want to do that anymore.

The other day I typed rm -rf on the command prompt and Linux went rogue, deleting a bunch of files. I think Linux is a danger to the world.

I'm with you. During most of my career, I was considered very technical and a good coder, but I never really loved the slowness of it. I enjoyed designing solutions but found "putting them into code" tiresome because for me the fun is in actually solving the problem.

But it's clear that many people feel otherwise. Software engineering has many classes of tasks and the difference in opinions about AI seem to exist because different people like different aspects of it.

Great! I was looking forward to an unified AI service like we got unified messaging with MS Teams!!!

I like his videos. He's funny and I don't have time to read his sources... So I'll take copy and paste, he's providing a good service.

That was deliberate. I don't think a casual reader, like most are, would have understood the nuance and the post would have been another casualty of indignation culture. I got downvotes as it was!

It's just considered one of the best movies and screenplays of all time. My quote is a relevant line in said movie.

"For years I paid my people extra so they wouldn't do that kind of business. Somebody comes to them and says, 'I have powders; if you put up three, four thousand dollar investment, we can make fifty thousand distributing.' So they can't resist. I want to control it as a business, to keep it respectable. I don't want it near schools! I don't want it sold to children! That's an infamia."

When AI can write screenplays like this one, then it's game over.

I missed the "Why I left Google" emails! If you haven't been on HN too long, these were a nearly daily staple back in the day.

The thinking was: "Google is a dream job, nobody would be stupid enough to leave it". Yep, once a dream job.

We're getting these posts now for OpenAI and Anthropic, now including Google, but the reasoning is a lot more sinister.

While before it was "Yep, colleagues and free food were great but I wasn't entirely happy for minor reasons X, Y and Z", now it's "I had to leave because I could have destroyed humanity!"

1. Find a secure random number generator that outputs a number between 1 and 1000

2. Guess the next number

3. You'll be reliably wrong

Do you have "superhuman predictive ability" in this case?

Ed Zitron has a similar success rate, perhaps lower

It makes sense from the point of view of a giant corporation. If people are paying despite of recent price increases and declining quality, why not try to increase profits even further?

Microsoft should now release a brand new UI framework, this time designed and implemented end-to-end by AI:

Microsoft Visualization Copilot

If there's anyone from Microsoft here, let your bosses know I'll accept a Chief Prompting Officer role to help get it (and much more) out of the door.

I'm strongly against destroying books to train AI. That said, I'd be extremely interested in exploring the contents of Japanese books, which probably have a lot of stuff not available outside Japan, through a LLM.