teaearlgraycold

Definitely. For now I'm making friends with lots of very capable people - it's not as easy to do that outside of a hub like SF.

> I think the amount of protectionism we’ll see could be absolutely enormous because corporations aren’t capable of altruistic attitudes.

On some fronts they will be powerless to stop the changes. I look forward to drinking the tears of IP owners when breaking through their DRM and recovering their source code is just a matter of sharing a model ID, prompt, and hyper parameters. Reverse engineer NVidia's firmware and drivers. Decompile and port Nintendo's latest console's operating system. We may soon find ourselves in a world where tasks like that are achievable in a few days on consumer hardware.

I'm a rare individual that has consistently made choices to earn less money in order to have more enjoyment. I would clock fewer hours than they wanted me to during college for my internships, I left a FAANG to go back to startups, I quit a startup to work on volunteer work and try to bootstrap a business.

At multiple jobs I've noted that I felt the pay should have been less. Not so that the employers could take a bigger slice of the pie (in fact, my preference is to be in an anarchist co-op, but I understand that's a wild take and might be terrible in practice). My hope was that we could filter for people that genuinely want to do the work and take on fewer people that were there for the money. At the same time more money is more better. And I like having money. It's complicated.

As I've said many times before, I didn't come to Silicon Valley to get paid. I came here to make computers do cool things alongside other people. I need income to do that, but I try to make sure I'm in a situation where the income isn't something I need to think about all of the time. If you're doing mercenary work that's not possible.

Good advice. But most work computers I’ve used were handed to me fresh from the Apple Store and they requested I wipe them before returning at the end of my employment. Not everyone works in places with an IT department, device management systems, etc.

I suspect people are afraid of change and hide behind truths like “it’s not conscious”. It’s not. But that doesn’t mean it’s not a force to be reckoned with.

LLMs have genuinely modeled systems purely through training on natural language descriptions of those systems. It’s a massive lesson and an answer to the Chinese room thought experiment. I think that’s reasonable to label as intelligence and goes far beyond a stochastic parrot.

This makes me want to get a job just to be able to run this with my coworkers.

It's easy to train LLMs to decompile code. All you need to do is take the source code from existing training data and compile it. Then you swap the order and show it the compiled code, then the original, and judge it by how well it predicts the original.

Well for one thing it's presumably illegal to use illegally retained data in court.

Where’s the party? I just built something similar to your product for an SF-based non-profit (fully self-hosted on site) and it’s been extremely useful. Would love to get tips on how we can improve our setup.

At least for me the Claude plans seem like an incredible deal and I never hit my limit.

I think I caused this release because I just embedded a sizable corpus with the previous model two days ago! You’re welcome everyone.

> what are we (in the US) even building these super massive data centers for?

Partially to make investors think it's worth giving US companies a lot of money. Also, I think distillation is a significant part of why Chinese models perform as well as they do. I think that's completely fair play (OpenAI/Anthropic/Google/Meta stole a lot of their training data). But I would expect if US model developers stopped right now Chinese development would slow down.

US companies are clearing the path, others follow in their wake.

A lot of cloud platforms have terrible slow network storage. They also might need to compile the GPU kernels fresh as they might not have a persistent CUDA cache.

Sure, and I love to see people actually giving a shit and getting engaged locally. There are many things I see - tweets and video from protests - that make me think people should take that energy to more direct issues.

> With the talk of AI data centers' impact on the world, you'd think this would be 10x to 100x the amount of energy in order to get the effects they're using here.

Well there's also training to consider. But yeah, most of what I see online about datacenters is nonsense. People worried about water use as a top concern are misinformed.

Look, I absolutely get it if you're worried your boss is just waiting for the day to replace you with an LLM. Fucking get organized with your fellow laborers instead of getting distracted by datacenter water use. Call your politicians to rein in billionaires. If anything the ownership class is probably directing online discourse towards electricity and water user to keep people away from class-focused debates.

Weird

It was more of a “sorry, not sorry”

Frontier from 9 months ago? I don’t know about that. But it sure punches above its weights.

Sorry but it’s looking more and more like the top American labs won’t have any kind of moat.

I remember my friend’s dad saying proudly how his computer came with Microsoft Copilot, back when the branding had only existed for like 6 months at most.

I can believe that a lot of people, too many, are lonely. But most? Not sure about that.

Look, I love code. I have laser cut code snippets framed on my walls. I functions that I have shared online proudly multiple times. I have projects I have lovingly tended to for years just for the joy of making a computer do something strange.

But what’s far far more important is making people happy by producing good software. Ever since my very first SWE internship I’ve cared most about the user. I’m extremely bummed if the user isn’t happy with my work. I’m ecstatic if my work can improve their day. LLMs change how my work gets done and what it is that my effort goes towards. But the core reason to be here instructing computers hasn’t changed.

It just needs to be something where you can verify its information quickly enough.

I find mostly what I do is think about how to improve the product, what the user wants, and what information the LLM needs and how to get it into the context window.

You’re right. Why am I wasting time on here?

It will never be cheaper to put compute into orbit. And costs for AI are dropping like a rock here on Earth.

The entire industry, including its outputs that LLMs are trained on, hasn’t reconsidered what’s easy vs. hard or fast vs. slow. LLMs consistently recommend against code changes because they will take “a weekend”. No, Claude. You will do the work and it will take 20 minutes.

New layer of abstraction. Someone needs to understand how to build that old C library and reprogram that embedded controller, but it doesn't need to be me. Give someone infinite possibilities. Tell them they can have any software change they want. Everyone will have some good ideas. Some people will be head and shoulders above the rest. Being able to do anything (not that LLMs exactly deliver on that goal) doesn't mean you'll do something interesting. It doesn't mean you'll invent a product worth paying for, or impress many people, or make anyone particularly happy or inspired.

I can confirm this person is the owner of onionfutures.com

I think it might be appropriate if what we're talking about is a human connecting an LLM to something it can't natively access. Like something in the physical world - a piece of hardware, a different human, software the LLM can't access. As the human in the loop your role might be to mostly just shuffle information back and forth from this external resource. But depending on the situation there could still be quite a lot of interpretation and analysis performed by the human in the loop.