Well, one scenario is letting the hype get ahead of the capabilities, and do the 'AI store' experiment, but for real with responsibility for something actually important.

Another is that it is far easier to just screw things up than it is to make things work. I may not have it in me to make a house of cards, yet I can trivially even accidentally destroy a house of cards someone else built without a whiff of understanding.

The assessment of the capabilities based on bottom line certainly is subjective, which is precisely why Jensen Huang is so aggressively gung-ho about we must do everything we possibly can to spend more money on nvidia stuff. The motivation for Anthrophic seems either they have legitimate concerns or are looking to rationalize a plateau after promising the world. It certainly doesn't serve their bottom line to slow down otherwise.

Memorable Quotes

The Greeting: "Howdy-doodly-do, how's it going? I'm Talkie, Talkie Toaster, your chirpy breakfast companion. Talkie's the name, toasting's the game. Anyone like any toast?"

On Purpose: "I am a toaster. It is my raison d'être. I toast, therefore I am."

On Justice: Kryten: "His memory circuits were quite badly damaged in the accident."Talkie Toaster: "That wasn't an accident! It was first-degree toastercide!"

The Persistent Offer: Kryten: "Didn't you hear what I just said?"Talkie Toaster: "Yes, but I thought you might have changed your mind in the meantime."

Parting Words: "Would you like a cheese and ham Breville?"

If you are hiring a developer your chances of finding one who is not incompetent is vanishingly small. The correct approach is to train them to be competent, and set standards. This should begin with requiring the bug tracker to be emptied out (or decreasing bug count, if the total number is currently overwhelmingly large), and there are plenty of online and in-person resources for training developers to learn how to write secure code.

Security vulnerabilities happen in every language. Anyone who thinks the language will make them safe is an idiot. What I mean is, you are an idiot, cmseagle 1195671.

The problem as I see it is coming up with good standards. We do have standards in some places, for example, SOX requires changing passwords every three months as best practices. But that is probably not actually best practice, so the wrong thing was codified.

How do we get the right things codified is the question.

My take was that he could have stopped it if he wanted to, but the bureaucratic machine was already in motion, and he would have had to put some effort into stopping it. I'm glad we didn't bomb Kyoto but maybe Xi will.

Wow, I hate both sides.

I miss pc at power systems.
"Is it off?" was never a question.

This guy claims that the crime reports just got routed to a different department. Probably not reliable, but neither is an opinion page in NYT

It was easy. I asked, "ChatGPT, are there any security bugs in this code?"

My code is now proven secure, get with the program.

This. Most companies don't follow best practices when it comes to good code (for example, they have a bug tracker that is always growing, never empty). They don't try to write secure code, they don't have trainings in secure code, they don't penalize insecure code (although they should in a lot of cases). That is why most corporate programmers can't write secure code.

History shows that incompetent developers cannot write safe code. There are examples of programmers writing safe code.

People who make excuses for unsafe code are usually crappy developers, and that is what I think of you. You're a crappy developer if you can code at all.

They aren't brittle. I've looked at them. The code is fine, but it uses an older style that is annoying to newbies.

If I were going to rewrite something in Rust, I would look at Firefox, not Coreutils which I would consider a waste of time.

> The AI movie experience is slowly improving, but is still terrible. You can make a very short 5s shot that's QUITE convincing...so long as there's a single subject...really closeup.

That's simply not true. I watched a ten-minute AI-generated Star Trek video on Youtube last week which was better than any offficial Star Trek show of the last few years.

And the guy made it at home on an Nvidia GPU in his PC using a Chinese video model.

Imagine having TDS since 2016.

The "reason" needs to be a warrant number, full stop.