Hacker Newsnew | past | comments | ask | show | jobs | submit | slashdave's commentslogin

We're social animals. Anything that benefits the extended tribal unit is advantageous. Adults beyond child-bearing age contribute significantly to child raising, education, leadership etc of the entire tribe.

Similar effects are seen in other species

https://www.nature.com/articles/s41467-018-05515-8


Failures are not publicly reported, in general. Do you we know what they have invested in?

Huh, yeah, the sky is blue outside and the sun is shining.

Although, I am feeling a bit lazy so let me see if I can simulate a walk.


Will join you in a bit, let me just pause the simulation to adjust the settings for "knee pain when walking" and "need coffee to jump start day"

This is a video model, not a world model. Start learning on this, and cascading errors will inevitably creep into all downstream products.

You cannot invent data.


Related: https://arxiv.org/abs/2601.03220

This is a paper that recently got popular ish and discusses the counter to your viewpoint.

> Paradox 1: Information cannot be increased by deterministic processes. For both Shannon entropy and Kolmogorov complexity, deterministic transformations cannot meaningfully increase the information content of an object. And yet, we use pseudorandom number generators to produce randomness, synthetic data improves model capabilities, mathematicians can derive new knowledge by reasoning from axioms without external information, dynamical systems produce emergent phenomena, and self-play loops like AlphaZero learn sophisticated strategies from games

In theory yes, something like the rules of chess should be enough for these mythical perfect reasoners that show up in math riddles to deduce everything that *can* be known about the game. And similarly a math textbook is no more interesting than a book with the words true and false and a bunch of true => true statements in it.

But I don't think this is the case in practice. There is something about rolling things out and leveraging the results you see that seems to have useful information in it even if the roll out is fully characterizable.


Interesting paper, thanks! But, the authors escape the three paradoxes they present by introducing training limits (compute, factorization, distribution). Kind of a different problem here.

What I object to are the "scaling maximalists" who believe that if enough training data were available, that complicated concepts like a world model will just spontaneously emerge during training. To then pile on synthetic data from a general-purpose generative model as a solution to the lack of training data becomes even more untenable.


How is it not a world model? The latents of the model apparently encode enough information to represent a semi-consistent interactuable world. Seems enough world-modely to me.

Besides, we already know that agents can be trained with these world models successfully. See[1]:

> By learning behaviors in imagination, Dreamer 4 is the first agent to obtain diamonds in Minecraft purely from offline data, without environment interaction. Our work provides a scalable recipe for imagination training, marking a step towards intelligent agents

[1] https://arxiv.org/pdf/2509.24527


Given that the video is fully interactive and lets you move around (in a “world” if you will) I don’t think it’s a stretch to call it a world model. It must have at least some notion of physics, cause and effect, etc etc in order to achieve what it does.

No, it actually needs none of that.

How would it do what it does without those things?

Like all these models work, by simple interpolation.

But how does it interpolate?

Pixel by pixel, time-slice by time-slice, in a 2D+T convolution. You provide enough examples of videos of changing point-of-view, and the model reproduces what it is given.

Yes, it reproduces what it is given by modelling the rules of physics, geometry, etc.

For example, image generators like stable diffusion carry strong representations of depth and geometry, such that performant depth estimation models can be built out of them with minimal retraining. This continues to be true for video generation models.

Early work on the subject: https://arxiv.org/pdf/2409.09144


They have a feature where you can take a photo and create a world from that.

If instead of a photo you have a video feed, this is one step closer to implementing subjective experience.


It's not a subjective experience. It's the mimicry of a subjective experience.

I get the intent, but moving to linux for better bluetooth support is... an interesting take

BT can be a shitshow on any OS. Some combinations work flawlessly, some don't. It's not even the OS' fault IMHO, but the device manufacturers'. My Bose headphones have the same problems under both Windows 10 and Linux.

How so? Bluetooth has been working out of the box (no tinkering) for me under Linux for the past ten years now across multiple devices. Including stuff like APT-X and LDAC. All with proper OS integration (I use Gnome). What's the story on Windows?

Same here. The story for windows, IME, is that my work Logitech BT keyboard works fine, but neither my sony nor shure headphones work at all. Windows says connected, but then disconnects right away. On the same PC which dual-boots linux, they both work fine, with LDAC for the sony and apt-x hd for the shure.

At work, we have BT Jabra headsets. I specifically asked for a corded version, I hate the latency for calls. My windows-using colleagues, for some reason, love wearing a wireless headset and talking through the laptop microphone.


Well I won't be buying Dell again.

Not a PR person myself, but why use as an example a parody topic for a paper? Couldn't someone have invented something realistic to show? Or, heck, just get permission to show a real paper?

The example just reinforces the whole concept of LLM slop overwhelming preprint archives. I found it off-putting.


X11 was basically used everywhere when it was released

Why take the drastic step of switching to linux (a difficult endeavor) when you can simply turn off key uploading.

Why continue to use an operating system that’s adversarial towards you?

I will never understand this from software engineers/tech people in general. That demographic knows how technology works, and are equipped to see exactly where and how Microsoft is taking advantage of them, and how the relationship is all take and zero give from their end. These people are also in the strongest position to switch to Linux.

The only explanation that makes sense to me is that there's an element of irrationality to it. Apple has a well known cult, but Microsoft might have one that's more subtle? Or maybe it's a reverse thing where they hate Linux for some equally irrational reasons? That one is harder to understand because Linux is just a kernel, not a corporation with a specific identity or spokesperson (except maybe Torvalds, but afaik he's well-regarded by everyone)


Or maybe Windows just works better for their use-case? Did you consider that?

It's convenient to attribute the reason to irrationality or cult-like behavior than to actually accept the real reasons.

Who is being irrational?


Microsoft is known for regularly altering the deal. Just because you configure the OS to not upload keys today, does not mean that setting will be respected in the future.

Pray I don’t alter it further.

Because that gives you a lot more control over your computer than just solving this particular issue. If you care about privacy it's definitely a good idea.

Because Microsoft absolutely will make it mandatory somewhere in the not so distant future.

you've baked in an unfounded assumption that bitlocker is even initially enabled intentionally by someone who knows that's a choice they can make:

> Here's what happens on your Dell computer:

> BitLocker turns on automatically when you first set up Windows 10 or Windows 11

> It works quietly in the background, you won't notice it's there

> Your computer creates a special recovery key (like a backup password) that's saved to your Microsoft account

> You might be reading this article because:

> Your computer is asking for a BitLocker recovery key

...such as after your laptop resets its tpm randomly which is often the first time many people learn their disk is encrypted and that there's a corresponding recovery key in their microsoft account for the data they are now unexpectedly locked out of.

https://www.dell.com/support/kbdoc/el-gr/000124701/automatic...


oh man, it's so difficult even teenagers can do it within an hour and all they have to do is click on a few buttons.

Yeah, the real question is what comes after the install...

Test which version? Part of the problem is that string theory is a meta theory. There will always be ways to escape any negative experimental result.


String theory isn't a theory it's a family of related theories sharing some common mathematical tools.

People talk about this as though it's an attempt at deception, whereas two people notionally working in string theory could in fact be proposing highly incompatible models which would be conclusively ruled out (and a lot of them have been in so far as that can be done - i.e. experimentation has put tight bounds on their possible parameters).


> There will always be ways to escape any negative experimental result.

Yeah... except that hasn't been proven. That's just your belief, right?

I don't think anyone has proven that string theory can yield no testable predictions. I think if someone had, that'd be a big deal.

And I don't think we should pretend that an open problem is closed just because we don't like it.


Provide a viable test, and you will be sure that an experimentalist will jump at the chance


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: