rocmcd

Self-awareness and consciousness would be hard to make a case for. Those terms are themselves fuzzy, but I have not heard a compelling case that would make it obvious LLMs have those traits.

I'm so confused why they didn't make this a 2-in-1. I would love to see an Apple iteration on a Surface Book or Yoga.

It's actually the remote machine that isn't working, but I get the frustration. What I did to fix this is to override the environment variables in your SSH config like this:

Host * SetEnv TERM=xterm-256color

Then how could Musk limit Starshield capabilities if the US military has control?

As things stand today, if it is running on a computer then it is indeed "just software," regardless of how impressive it may be.

If we get to the point where we could emulate a brain down to the atomic level, then I may feel differently. That's not what we are doing today, though.

Does treating them as pets make for a better argument? Pets have agency and can behave in unpredictable ways. If my pet damages someone else's property then I am held accountable. I may not have foreseen how my pet could have caused said damage, yet I am still held accountable.

LLMs are extremely impressive pieces of software, however they are still just software. OpenAI's software hacked another company. The engineers may not have intended for their software to specifically take the actions leading to that outcome, but it was ultimately still their software. Lack of intention doesn't mean there wasn't negligence.

Same here. The video looks cool, but what I see is much messier and hard to identify shapes and objects. Tried on latest Brave and Firefox on Arch Linux.

I find it ironic that between the money they have raised, the talent they have in-house, and the quasi-super intelligence they have at their disposal, you would think they could have just created their own brand of smartphone (or done something like One Plus / Nothing).

It's hard to read Gwern's accompanying article without seeing this for what it is, which is a kind of mania.

I'm sure he means well and is genuine in his aspirations, but what's outlined for GA is framing LLM's as quasi-gods, which they absolutely are not. I wish him the best, and look forward to being proven wrong.

If you didn't have the time to write something, why would you expect someone to spend the time to read it? Did _you_ read all of the documentation you generated?

"Hyperactive and distracted" is not necessarily the exact reason, but there is a large, well documented gap in performance for boys vs girls in elementary school (at least in the US).

Well put. I agree wholeheartedly with your sentiment.

Maybe this is me just being angry at the new world that's being created, but the beauty of the open source ecosystem was humans giving away things they found useful in the hope that other humans could find them useful too. Having a machine take all of that and regurgitate it somewhere else without that connection (for profit, no less) feels like a betrayal of that open source ethos.

Now in the back of my mind I worry that everything I open source will be scooped up by corporations to make them more rich and more powerful, so I end up not publishing anything (not that it was of any value). I suspect I'm not alone in feeling that way.

Maybe not every 12 months / session, but I definitely think having a way to "sunset" certain types of laws would be a net positive on society.

Lovely. Thanks for sharing.

Are there directories for similar items? I love the idea of a simple, single purpose device that has maintenance and repair-ability at the forefront.

You're getting downvoted, but this is really the only answer here. Companies won't stop acting this way as long as their shitty behavior is rewarded, and people keep rewarding their shitty behavior.

No amount of legislation is going to prevent them from doing this. This settlement even proves that they can keep doing it with impunity!

Unless you have worked with Oracle or other big enterprises, you may not realize the scale of how these companies operate and the breadth of what they actually do. Just by looking at their product page[0] you can see they offer software, hardware, cloud, consulting, support, and even financing solutions. In addition to the technology and product people (of which there are many), you also need HR, sales, marketing, accounting, support, etc for the entire global organization.

Sure, 100,000 people is a lot, but Oracle also does a lot.

[0] https://www.oracle.com/products/

One thing that I think goes under discussed when it comes to the big AI companies is the _insane_ transfer of knowledge from sources like Britannica (among others) to the AI companies, which now use that knowledge to make money.

If deemed legal, what incentive is there to create and share these datasets if some company can come and scoop it up and make money from it? All without attribution, copyright, or sharing any of the revenue.

Regardless of illegality, it's a complete violation of trust and effectively poisons the well of the open internet.

I couldn't agree more. I think what has happened is that everything has become so fragmented due to the sheer volume of content being created that discovery is still a challenge. If you are someone who thinks that music, movies, novels, etc are in decline, then frankly you're just not looking hard enough.

For music, I'd recommend looking at Bandcamp Daily[0]. It's not all my cup of tea, but there are some amazing new artists out there spanning nearly every genre imaginable.

[0] https://daily.bandcamp.com/

As a parent, this is a hilarious comment.

This is like saying "why don't you just teach your cat to use the toilet instead of using a litter box?". I mean, yeah, that sounds awesome. Given infinite time and energy, I'm sure it's possible. Best of luck to you, though.

And I don't say that to be rude or disparaging, it's just that parenting is a little like war: your plans never survive contact with the enemy. I had similar thoughts and ideas before I had kids, and they all went out the window when you deal with the real thing. Sleepless nights, a screaming infant, being scared out of your mind when they're sick... but then you will find a calmness unattainable anywhere else as you hold your sleeping child. All of your accomplishments will pale in comparison to the joys of parenthood, and you will unironically look back at those years as some of the best years of your life. You will see.

There's no need for (local) AI acceleration if you are leveraging a remote LLM (Claude, ChatGPT, etc). The vast, vast majority of users are most likely just making API calls to a remote service. No need for specialized or beefy hardware.

I agree with the sentiment, but that is completely unrelated to the topic at hand.

Just because Congress is stuck doesn't mean the Executive gets to do whatever they want.

You would think that if you genuinely wanted to curb illegal immigration, then this would be the way to do it. People come here for money. Take away the money and they will no longer come.

Hell, you would probably have bipartisan support for nationwide crackdowns on employers who are employing anyone here illegally. They are undercutting American employees and dodging taxes. Who wouldn't be for that kind of law enforcement?

Instead we get unaccountable masked men with guns murdering citizens and terrorizing an entire populace. Imagine if an "ICE raid" meant a team of accountants showed up at a business and gave them a hefty fine for employing anyone here illegally. It seems like that would be much more effective, which makes me genuinely wonder if the demonstration of strength through cruelty that we currently have hasn't been the goal all along.

Yeah, I'm not aware of any regulation preventing new painters/drywallers from entering the market that would drive up costs.

This is a pretty straightforward example of the Baumol effect, where _anything_ bespoke (not manufactured) requiring a human is simply going to cost more. The materials for patching drywall/plaster are tiny, it's the cost of the person that is expensive because overall cost of living is rising. The cost of outsourced labor (which you can leverage when making a TV, but can't for local labor) also probably plays a role.

In fact, I bet you could find someone to fix the drywall/plaster much cheaper than the cost of a TV. You just won't like the quality of the work.

I don't doubt that they have improved a lot this year, but the same claims were being made last year as well. And the year before that. I still haven't seen anything that proves to me that people are truly that much more productive. They certainly _feel_ more productive, though.

Hell, the GP spent more than $50,000 this year on API calls alone and the results are... what again? Where is the innovation? Where are the tools that wouldn't have been possible to build pre-ChatGPT?

I'm constantly reminded of the Feynman quote: "The first principle is that you must not fool yourself, and you are the easiest person to fool."

This has been my biggest question through this whole AI craze. If AI is making everyone X% more productive, where is the proof? Shouldn't we expect to see new startups now able to compete with large enterprises? Shouldn't we be seeing new amazing apps? New features being added? Shouldn't bugs be a thing of the past? Shouldn't uptime be a solved problem by now?

I look around and everything seems to be... the same? Apart from the availability of these AI tools, what has meaningfully changed since 2020?

QA? What's that? /s

In all seriousness, it's been a long time since I've seen a dedicated QA position instead of just assuming that devs will test as they go.

The mummy example is interesting, but we're not trying to reanimate mummies in an artificial way so I'm not sure it poses the same issue. There's also an educational and historical aspect with mummies that is missing here.

This also allows for the creation of a kind of "anti-history", where someone could create a fictional history for nefarious purposes.

In fact, a common undercurrent I see with many AI use cases is accelerating the breakdown of trust in society. You can no longer trust anything you read, see, or hear.

Maybe I'm getting old, but, I have to say, seeing stuff like "Alan Turing™" (a legend and titan of computing history, followed by a ™) really bums me out. I don't believe for one second that any of the deceased people listed here would want their likeness emulated in such a hollow, mechanical, and inhuman way for all of eternity.

Exactly, they wouldn't even need all of the emails in gmail for that example, just the ones from a specific account.

The real equivalent here would be if gmail itself was injecting NYT articles into your emails. I'm assuming in that scenario most people would see it as straightforward that gmail was infringing NYT content.