timmytokyo

Let me guess. You also believe grok's recent episode, where it started inserting "white genocide" into the responses of totally unrelated queries, was caused by a rogue employee totally not doing it at Elon's behest. Despite the fact that Elon is always going on about "white genocide".

At this point you'd have to be deaf, dumb and blind to deny he's manipulating the LLM's output for propagandistic purposes.

What's to check? Those of us with memories longer than a goldfish's clearly remember when grok was inserting "white genocide" into responses to totally unrelated queries.

A veritable "explosion"! Surely GP can name 3.

Perhaps he felt a jail cell would offer him a better life than what he was getting in society.

Yes.

It's certainly a choice to lead with the funding and a photo with a venture capitalist.

In a healthy economy there would be hiring in other industries, helping to offset those numbers.

Yes and no. LLMs take all the writing on the Internet (good and bad) and average it out. It's similar to the way generative AI images always have an identifiable, artificial "look". They've averaged out the personality and thereby erased the individuality that went into the efforts the original artists used to create them.

It's called the Big Lie. Tell an egregious lie often enough, and some people start to believe it.

Accountability is for suckers, I guess.

Exactly. Programming skill, like most human skills, is acquired through practice and repitition. You can't become an excellent programmer by observing other people (or LLMs) doing it. That's like arguing you can become a master musician by reading sheet music or watching live performances by other musicians.

How coincidental that I was just reading something related [0] before seeing this post.

"Silicon Valley is bad at politics. If nothing else during Trump 2.0, I think we’ve learned that Silicon Valley doesn’t exactly have its finger on the pulse of the American public. It’s insular, it’s very, very, very, very rich. [...] I expect it to play its hand in a way that any rich 'degen' on a poker winning streak would: overconfidently and badly."

And...

“People don’t take guillotines seriously. But historically, when a tiny group gains a huge amount of power and makes life-altering decisions for a vast number of people, the minority gets actually, for real, killed.”

[0] https://substack.com/home/post/p-187592016

Nate Silver often annoys the hell out of me, but I think he's right about some of the possible political impacts of AI.

"Guns don't kill people, people kill people."

You're glossing over January 6. There are few things more damning of the American electorate than their willingness to vote for a man who tried to use a coup to stay in power. The rest of the world sees it. And they'd be stupid to trust that Americans wouldn't do something so stupid again.

Assange's lawsuit is kind of silly, but his point about the incorrectness of the award to Machado stands up to scrutiny. She overtly encouraged military intervention by the US in Venezuela. That's a blatant contradiction of everything the Nobel Peace Prize is purported to stand for.

Replace "Claude Code" or "AI" with "Jesus". It all sounds very familiar.

>I am not aware of an easy way to load (managed) mods as DLLs to IL2CPP-compiled game. I am thinking about `Assembly.LoadFrom("Mod.dll")`.

Ah, I was thinking native DLLs (which is what we're using on a project I'm working on). I think you're right that it's impossible for an IL2CPP-built player to interoperate with a managed (Mono) DLL.

>If you have some suggestions [re: serialization], I'd be interested to see what options are out there for C#.

We wrote a custom, garbage-free JSON serializer/deserializer that uses a fluent API style. We also explored a custom codegen solution (similar to FlatBuffers or protobuf) but abandoned it because the expected perf (and ergonomic) benefits would have been minor. The trickiest part with Unity codegen is generating code that creates little to no garbage.

>We don't use IL2CPP because we use many features that are not compatible with it. For example DLC and mods loading at runtime via DLLs, reflection for custom serialization, things like [FieldOffset] for efficient struct packing and for GPU communication, etc.

FieldOffset is supported by IL2CPP at compile time [0]. You can also install new DLLs and force the player to restart if you want downloadable mod support.

It's true that you can't do reflection for serialization, but there are better, more performant alternatives for that use case, in my experience.

[0] https://docs.unity3d.com/Manual/scripting-restrictions.html

We should just develop cold fusion. It's gotta be easy, right?

This response totally misses the 500 billion dollar elephant in the room.

It's a tell that he's been influenced by rationalist AI doomer gurus. And a good sign that the rest of his AI opinions should be dismissed.

If only they had discovered philosophy. Instead they NIH their own philosophy, falling into the same ditches real philosophers climbed out of centuries ago.

Not sure why you're being downvoted. It's such a common phenomenon that it has its own name: Nobelitis.

[0] https://en.wikipedia.org/wiki/Nobel_disease

From the article:

'Clement said that hotels limit speeds to restrict peer-to-peer downloading, and suggested that universities do the same. “I don’t think it would be the end of the world if universities provided service at a speed that was sufficient for most other purposes but didn’t allow the students to take full advantage of BitTorrent,” he said. “I could live in that world."'

Insane. So Sony's lawyer is arguing that every university student should have sub-broadband internet speeds simply because a small fraction of the students infringes copyright.

I see very little evidence for this abundance of American "common wisdom". If anything, America has always had a deeply anti-intellectual vein running through it, whether it was the Scopes monkey trial, the Know Nothing Party or what you see in the present political scene. Higher education, especially the affordable kind at public universities, has been a bulwark against the paranoid delusions that often dominate societies that revere superstition and "common sense" over reason, empiricism and humanism.

When building their LLMs, the model makers consumed the entire internet. This allowed the models to improve exponentially fast. But there's no more internet to consume. Yes, new data is being generated, but not at anywhere near the rate the models were growing in capability just a year ago. That's why we're seeing diminishing returns when comparing, say, GPT-5 to GPT-4.

The AI marketers, accelerationists and doomers may seem to be different from one another, but the one thing they have in common is their adherence to an extrapolationist fallacy. They've been treating the explosion of LLM capabilities as a promise of future growth and capability, when in fact it's all an illusion. Nothing achieves indefinite exponential growth. Everything hits a wall.

As if inference exists in a bubble. Driving a car from point A to point B costs $0, as long as you exclude the cost of the car or the fuel you purchased before you were at point A.

Heaven forbid that anyone discover the horrible things our tech overlords are doing in the world.

I think it's because most software developers have experience working with assholes. Commenting on someone famous who's being an asshole is a natural way to vent.

If you've worked in SWE for long enough, you'll run into this kind of socially maladjusted petty tyrant many times in your career. It's fascinating to see so many of these tyrants exposing themselves in this thread, though.