icapybara

I think one big difference might be that we are limited by hardware. For this to be true, the models and algorithms will need to become exponentially more efficient over time, because the hardware won't.

> If something goes wrong – maybe it fails or yields biased results – you can’t just blame a third-party developer of the technology.

Implicit shoutout to all the “slap a front end on ChatGPT” products out there.

> The institution produced a fair amount of academic research, but the vast majority of it was essentially fraudulent. We had some contracts with industry to produce technology in the space, but these mostly seemed to consist of the organisation fumbling the projects and massaging the truth (read: lying) until they could collect more funding.

> Piles of money + unclear outcomes = every grifter under the sun begins to migrate to your organisation. It is very hard to keep them all out, and they naturally begin to let other grifters in because they all run interference for each other.

Wow, this is very honest, which is refreshing in today's world. Thanks for the post OP.

Why is element-wise multiplication for a multi-dimensional array a "footgun"?

You try it once, realize it's doing things element-wise, and then go lookup the proper product symbol (@ in numpy).

I don't understand why Reddit (or some other big name) didn't capitalize on the moment and put out a quick Twitter clone.

ChatGPT is already ChatGPT for generating prayers.

They all know this is how science works. It's nonetheless surprising. They're not shocked that they're wrong, they're shocked about what this could imply about the universe.

I see Hacker News is susceptible to the same anti-status-quo radicalism that Reddit and many other online forums are.

Now this is the future. We're living in the androids timeline.

Who can say for sure if the probability is over- or under-estimated? Nuclear war has only happened once.

Is anyone getting access to Bing Chat yet? How fast are they opening this up to users?

Why should they all do this?

Author says he was slow to get into AI because of his skepticism falling for the hype behind new technologies (like NFTs) but clearly also fell for the hype behind those technologies according to his past posts and Twitter.

It seems the author will jump on any tech bandwagon, so take this writing with a grain of salt.

(This isn’t a comment on AI, just on the credibility of the author)

Right, I mean, how could it know?

I could write in the tone of ChatGPT if I tried hard enough. It's an intractable problem and a tool like this probably does more harm than good.

Is it?

How do we know our own 'processes' could not be modeled that way?

I think this is the point of the Turing test. At some point, you can't tell if a system is "thinking" or just crunching numbers. And it doesn't matter.

I don’t understand products like this. If you don’t own the model, what value do you provide?

Based on some of the comments, maybe I'm "holding it wrong." I don't know, but I spent a good 20-30 minutes trying to get this sorted out before deciding it wasn't worth the time.

Love Firefox, and I like what the Multi-Account Containers are trying to be, but it just didn't work for me. And it sounds like I'm not the only one who's run into this issue.

I tried to like this the other day but it was a real hassle.

I have 3 accounts on a website. I thought I could setup 3 different containers, one for each account. But it turns out that switching between containers is not simple. Whenever I tried to switch, I got a pop up from Firefox asking me if I was really sure I wanted to switch my default container for that website. Sometimes it sent me into a redirect loop with the prompt also. I did some googling and found that the consensus was that containers simply aren’t meant to be used that way.

Firefox Multi Account Containers seem to assume you will always use one container for one website, and you can’t easily use multiple containers for a single site. Or at least, it’s very inconvenient to do so.

In what ways do you view working for OpenAI as a bad thing?

We got better at detecting and recording them.

That's true except also every other point in space is at the center of its own balloon. The balloon has no edge, the movies just become too dark to see past a certain distance.

It didn't start somewhere. The Big Bang happened at all points in space. It happened "everywhere".

Or maybe they couldn’t because their early galaxies lacked heavy elements. Or anything FTL is simply impossible.

Interesting. I hope OpenAI is charging them for all that value they’re skimming off of OpenAI’s model.

Maybe a dumb question, but how do you train a model that has already been trained? What is that process like?

Nice analysis. I think the author really hit on the key problem: expensive BOM due to poor engineering choices.

They also face a more competitive environment now than possibly ever before. There are other game stores/fronts like Epic's and Activision's, and the console giants are coming for a piece of their pie (Xbox Game pass and Sony's equivalent). There's also the ever-looming risk that Apple decides to go after their market with Apple TV/iOS somehow.

If I was an executive at Valve, I'd probably be happy with the current success, but nervous for the future.

I don’t have statistics. It’s in the name though, “silicon” valley. The South Bay is home to the semiconductor industry and its ecosystem.

My own company has about 15k employees in San Jose, and we’re all back to the office full time. Many of us can’t do our jobs from home, it’s hardware work.

Much of Silicon Valley is hardware work that literally can’t be done from home.

Oh. Misread that. Nevermind then.

Sound waves (phonons) are probably worse