drakythe

Nothing solid. All the reporting I see is that numbers are measurable but not on the level many people expected, and that these systems are proving incapable of wholesale replacing employees, which clicks with my own knowledge of some large layoffs accompanied by press releases about streamlining corporate structures using AI and the outgoing employees asking “what AI?” Because they didn’t train an offshore team or document their department responsibilities before being let go.

My suspicion (and many others as well) is that AI is used as a way to cover for large corporate restructuring in the face of money no longer being available for “free” due to interest rates.

What would a decision model add to and audio transcriber?

Humans can be sanctioned and fined, and eventually disbarred if they continue to lie in court filings.

With regard to certain legal questions this has always been the case. AT&T Fought the US Government for 20 years and eventually won because the government gave up. Without some kind of national anti-SLAPP law we're all one irritated oligarch away from having our lives financially ruined.

I am curious what level of trust established law firms treat LLMs with.

a note about "Doom on a preganancy test". It wasn't actually running on the pregnancy test. It was displayed on the tests 1 color LCD screen, but everything else was different hardware. It was a test of hacking the display and what it was capable of in cheap disposable device, not the processing power of the electronics in the test.

I am also interesting knowing how they determined the model was sandbagging rather than just making a poor decision.

Also, this paragraph makes me wonder about all their stats on the exploitation and misalignment charts. If the model is that good at hiding "incriminating information" and sandbagging, are they sure its alignment is that?

Pi * r2 (squared) both pans. Divide smaller pan area by larger pan, now you have the % of how much the smaller pan recipe fills up the larger pan, and the missing % you need to fill. Increase ingredients by that % divided by the filled %.

Small pan area: 20 sq cm Larger pan: 48 sq cm

20 / 48 = .42, I'm missing .58 of the pan. .58 / .42 is 1.38. My recipe needs 2.38x the original to fill the larger pie pan.

Oh yes, absolutely. Meds are extremely useful and given the choice between "fight myself AND the system" or just "fight the system" I'll choose to stop fighting myself every time.

I think it is also worth reframing what "functioning well" means. Society expects certain things of us and our brains function in a way that those things are more difficult. Is that our fault, or society, or some mix? Isn't it possible that society itself could adjust in such a way that people who naturally struggle to sit in a cubicle doing the same thing all day every day could be offered other options? Seems like our capitalistic system that requires "productive" work as measured by those who already control the capital is maybe also a problem here.

I'm on stimulants and I agree that the post you responded to needed reframing. But I think we do ourselves and anyone else who isn't "normal" in body or mind a huge disservice just focusing on acceptance of treatment when there are larger societal changes that could be made.

My statement was perhaps too broad. It is possible for ADHD people to get high off their medications, yes. But that is a signal the dosage is too large. Normal dosage and usage won't get a person "high" in the traditional sense, much in the same way injecting insulin into a healthy person is bad but a diabetic person requires it for standard function, however too high of a dose would still be bad for a diabetic person too.

Suicidal ideation as a side effect is bad. I'm sorry you had that experience.

Its not addiction at all. I literally forget it once or twice a week on average, even on work days. I don't think you know what you're talking about.

Diagnosed with ADHD as an adult, and on stimulants (went through 3 kinds before we found one that seemed to help).

What "high"? Our brain chemistry literally works differently than neurotypical people and the stimulants don't get us high. In fact they often have a calming effect. I had to fight not to fall asleep at my desk two days this week (Not entirely the medication's fault, lotta family stress wrecking my sleep).

Does tolerance eventually cause the dosage to increase or the medication to cycle? Yes. But as others have mentioned taking the weekend off (or hell, just forgetting periodically in my case) mitigates that to a large extent. And, since there is such a hard time acquiring the medication even with a prescription skipping weekends or vacations has the benefit of seeing people through the inevitable shortage their pharmacy will have.

Apple's hypocrisy is a longstanding issue with regard to their apps vs 3rd party apps. As I recall when the Green Bubbles first happened in messages someone pointed out that Apples own design guidelines specified a minimum contrast in colors or they would reject apps, and the green bubbles with white text did not meet that threshold at the time, and it stayed that way for quite a while. And for all I know the light themed version of the app still does (I have mine on permanent dark theme).

https://medium.com/@krvoller/how-iphone-violates-apples-acce...

I'm curious who refused you the refund? I seem to recall it being refunded pretty steadily by everyone who sold it digitally. Maybe I am misremembering.

Still entirely fair to have been turned off by the experience. Plenty of developers or publishers I don't purchase from anymore due to similar shenanigans.

"A while back" -- One week ago.

I am curious if they are still implementing the process or if this particular attacker already figured out a way around it.

If you view the paper in Arxiv's experimental HTML mode the text is faintly visible, no custom styling required.

No human souls imprinted in the LLMs (Yet), that I am aware of. I think we're safe for now.

But I'm absolutely joining the Machine Crusade if we have a first Impact event and I survive. Some days I wonder just how flabbergasted Frank Herbert and other pioneers of sci-fi would be at the situation we find ourselves in today.

It is interesting, even the most random LLM model announcement gets plenty of discussion about the merits of the model or others, or the harness, or something. This one has like... 2 comments about the announcement itself and the rest of meta commentary on the announcement being clearly LLM written and availability.

I really hate the constant use of bold+underlined text in places that are not links. Other underlined text are links, but bold+underlined is just emphasis. Feels like terrible UX, especially since they use it on claims that I would love to see the source of.

I want to sit down with a sales person who doesn't work in tech day to day and see if they can recognize this kinda thing or if its only people looking for it and who have experience in the domain.

Chiming in to say that health insurance and benefits are a huge part of what I look for in job listings, and all of my friends currently looking for new jobs too. It may feel less relevant when we were young, but approaching middle age and it becomes a huge driver of continuing in a shitty job and in accepting an offer.

Interesting too that they would include "ai friend" when China just added restrictions on AI "partners" such that many services stopped offering them rather than try to adhere to the restrictions.

No one. I do not believe this to be a good idea. I think its pretty awful and this whole thing, if it goes as far as they keep saying it is going to go, will spin wildly out of control and do an immense amount of damage to people.

But the big money has been skipping Leg Day for ages, and that lack of a foundation will bite us all. Which sucks.

Sunk cost fallacy.

So much money has been invested in Anthropic and OpenAI at this point that to declare it a loss and walk away could potentially destroy a lot of VC firms, and a non-trivial chunk of the US Economy.

Our current incarnation of capitalism is all about monopolistic behaviors. If these LLM companies do get to the point of being able to replace employees I fully expect them to stop selling shovels and start producing the gold directly, anyone else be damned. And, frankly, this has always been the case. If a product is built on top of another service it has a limited lifespan. Either the product will be purchased or it will be replaced by the service it depends on. How many "killer apps" has Apple silently absorbed into iOS over the years?

Dev apps need to be re-loaded every week (or two?) last time I did sideloading. The idea behind a dev app, in Apple's mind, is that it is for limited testing, so they have an expiration when not signed/installed through the App store.

Ah. I've never used Grok so I was assuming that meant there was actually a repository on the user's home directory, something I know is pretty common.

And this is why so many people run these inside of VMs. Still baffles me how these tools became so accepted when tossing out a `curl -o example.com/script.sh | bash` would be met with (rightful) skepticism until that script was examined.

I'm not seeing the information about it having been run at $HOME, where are you seeing that?

Doing all of that math and tracking them is a huge order. Outer Wilds famously simulates an entire solar system using Unity and they had an issue early in development where bugs would occur more frequently as the player visited the edge of the solar system due to floating point math getting wonky in the engine and the solar system's Sun being 0,0,0 coordinates. Their solution? Make the player coordinate 0,0,0 and everything else moves _around_ the player. That's right, in Outer Wilds, when you jump, the planet you're on is actually moving away from you. But they managed to use this method to simulate newtonian physics pretty well.

And even that solution is only temporary. Its possible to watch the simulation go on so long that planets begin to de-orbit the sun as the math simulation breaks down. For spoiler reasons players don't run into this issue, but it exists.

PS: If you haven't played Outer Wilds and you enjoy exploration/puzzle games go play it. Avoid spoilers if possible.