matheusd

Unfortunately, it doesn't look like this is sufficient.

While I had great success with GrapheneOS in the past, bank apps in Brazil have started blocking it, even when the profile you run it under has Google services installed. So GrapheneOS (again, even with all Google Play Services and all other dependencies installed in a given profile) is still not completely transparent to apps.

This may be a coincidence (as I don't use it every day), but I noticed blocking started just as the recent Felca Law (which introduced mandatory age verification for every software, app and OS in Brazil) came into effect.

How about this for an evaluation: Have this (trained-on-older-corpus) LLM propose experiments. We "play the role of nature" and inform it of the results of the experiments. It can then try to deduce the natural laws.

If we did this (to a good enough level of detail), would it be able to derive relativity? How large of an AI model would it have to be to successfully derive relativity (if it only had access to everything published up to 1904)?

(Not op)

Maybe not you in particular, but I expect people to be more forthcoming in their writing towards LLMs vs a raw google search.

For example, a search of "nice places to live in" vs "I'm considering moving from my current country because I think I'm being politically harassed and I want to find nice places to live that align with my ideology of X, Y, Z".

I do agree that, after collecting enough search datapoints, one could piece together the second sentence from the first, and that this is more akin to a new instance of an already existing issue.

It's just that, by default I expect more information to be obtainable, more easily, from what people write to an LLM vs a search box.

Explains why RL helps. Base models never see their own outputs so they can't learn "this concept exists but I can't actually say it."

Say "Neuromancer" to the statue, that should set it free.

Humans are more efficient watt for watt than any AI ever invented.

Indeed they are. For now. The long term trend is not in our favor.

block cars in cities a couple of days per week, individually selected per person.

The net result in São Paulo (Brazil) for (something that approaches) this is that people end up buying a second vehicle.

Local governments in BR have already made ads using generative AI that were shown during prime time TV hours[1].

You can argue that is a bad thing (local designers/content producers/actors/etc lost revenue, while the money was sent to $BigTech) or that this was a good thing (lower cost to make ad means taxpayer money saved, paying $BigTech has lower chance of corruption vs hiring local marketing firm - which is very common here).

[1]https://www.cnnbrasil.com.br/tecnologia/video-feito-com-inte...

in Brazil it is 20% (sugar cane based)

30% in 2025 for cars (from 27%), 15% biodiesel in diesel for trucks (from 14%).

Source: https://www.em.com.br/politica/2025/06/7183470-governo-aumen...

The whole bio-fuel industry is a very complex mix of economics (often requires subsidies to make sense), geopolitical (less imported oil), environmental concerns (mass scale farming soil degradation and CO2 emissions derived from it) and logistical (completely different transportation and refining process).

Don't forget lobyying by the relevant sectors!

"has existed long before humanity" isn't relevant for my argument.

"Will exist long after humanity" -> maybe, maybe not. If we're smart, capable and humble enough, we could, in principle, intentionally outlast them.

By "intentionally" I mean: we can design our future lightcone such that, by whichever measure you care to choose, there are still humans around. Yes, bacteria could be still around, but it won't be because they _chose_ to be around, it will be because it just so happened that the universe arranged itself in a way that they are still around.

By "in principle" I mean: if we spent enough resources, energy and smarts and built a civilization around this goal, we could plausibly (given the known laws of physics) do this. Whether we _will_ do it or destroy ourselves first any of the possible various means, is an open question.

Lineages of bacteria that exist today, here, will only keep existing in the _far_ future (billions of years from now, after the sun chars Earth and then spends its energy budget) if it just so happens that a panspermia event kicked some off our solar system and then they just so happen to find a suitable solar system to keep existing.

We can design our future, bacteria can't.

Pretty sure no bacteria will survive on earth after the sun expands enough to char it, yes.

Even if I'm wrong, and it does survive _that_, then it eventually won't survive the sun spending its entire energy budget.

We're the only ones that could intentionally (as in, actively design our future lightcone) to survive that, so that makes us special in my book.

Choosing what traits one considers exceptional, will by definition select what species one considers exceptional.

/Every/ other species that has /ever/ lived will cease to exist (at the latest, in a billion years or so).

Humans are the only ones (so far, anyway) that have any hope of surviving more than that.

That seems pretty exceptional to me :P

Disclaimer: the fact that we're exceptional doesn't mean we don't do dumb things and we shouldn't improve and do better.

You're right to doubt it!

And indeed, accounting for externalities (unmeasured or unmeasurable) is a tough economic proposition. If it weren't hard to account for every single variable, creating a planned economy would be easier (ish).

FWIW, there's a whole sub-field just dedicated to determining the value of life for various purposes (a starting link: https://en.wikipedia.org/wiki/Value_of_life). You may disagree with any specific assessment, but then you have to argue how that value should be calculated differently.

The counterpoint is that not doing so (implying some sort of infinite monetary loss if the entire human species is wiped out) would mean you want to spend every single unit of monetary value of the entire global economy to preventing this (which is also obviously nonsense - people have to eat after all).

So you have to put the monetary value somewhere (although you're completely within your right to question this specific amount).

I don’t think they can yet self improve exponentially without human intuition yet

I agree: if they could, they would be doing it already.

Case in point: one of the first things done once ChatGPT started getting popular was "auto-gpt"; roughly, let it loose and see what happens.

The same thing will happen to any accessible model in the future. Someone, somewhere will ask it to self-improve/make as much money as possible, with as little leashes as possible. Maybe even the labs themselves do that, as part of their post-training ops for new models.

Therefore, we can assume that if the existing models _could_ be doing that, they _would_ be doing that.

That doesn't say anything about new models released 6 months or 2 years from now.

Is it though? It is my understanding that the quantum fluctuations that give rise to BBs will still exist, even after (and specially after) the evaporation of black holes (perhaps assuming no Big Rip).

Attempting to summarize your argument (please let me know if I succeeded):

Because we can't compare human and LLM architectural substrates, LLMs will never surpass human-level performance on _all_ tasks that require applying intelligence?

If my summary is correct, then is there any hypothetical replacement for LLM (for example, LLM+robotics, LLMs with CoT, multi-modal LLMs, multi-modal generative AI systems, etc) which would cause you to then consider this argument invalid (i.e. for the replacement, it could, sometime replace humans for all tasks)?

Interesting, I have the exact polar opposite perception.

I accessed this through a Qubes AppVM (no GPU, limited memory and CPU budget) and the presence of videos makes this a very slow scrolling experience for me.

In general, anything that involves JS/CSS animation/blur/effects makes sites pretty slow (up to unusable for me). The unlogged homepage for github.com for example, spins my cpu at 100%.

If a generic human glances at an unfamiliar screen/wall/room, can they accurately, pixel-perfectly reconstruct every single element of it? Can they do it for every single screen they have seen in their entire lives?

Maybe it's a liability issue, not a competency issue.

Unless there is a small enough number of them that they can coordinate and all follow the same strategy.

That ("_every_ money-making industry...") seems like a too strong statement and can be proven false by finding even a single counter-example.

gwd's claim (AFAICT) is that _specifically_ OpenAI, _for this specific decision_ is not driven by profit, which is a much weaker claim. One evidence against it would be sama coming out and saying "we are disabling codex due to profit concerns". Another one would be credible inside information from a top-level exec/researcher responsible for this subproduct to come out and say that as well.

My favorite site re: end of the world/universe is the old timer http://exitmundi.nl

You cannot "generate evidence at will to meet any threshold of evidence" for this specific example ("all swans are white") because, unless you posit that there are swans in Kepler-422B, I can set my threshold of evidence to "we have turned the earth into computronium and have not found any non-white swans", which is a finite (albeit large) threshold.

Once we agree on whatever definition of "swan" we care about (an animal belonging to a specific phylogenetic branch, with such and such qualities) we can agree on a more reasonable threshold, one that should include a visit to Australia to be crossed.

You're describing certainty. This is a prediction, not reality.

I was being concrete about atoms, in particular, not about what gets named as a "scientific fact" in general.

Yes, in general you don't reach that threshold of being able to produce a stream of evidence that is always compatible to a theory, and you should not call that theory a fact (I also pointed that out on a different message). However for the existence of atoms, specifically, I do believe we have reached this level.

Certainty does have interesting properties, from a Bayesian perspective: if you are certain of something

I don't literally believe with 100% certainty that atoms exist (because I may as well be a Boltzmann Brain hallucinating my entire universe). I say there will never come a time when we'll disprove the existence of atoms (at their level of abstraction).

Value != Price, so yes a Toyota may be more valuable than a Lamborghini.

A Lamborghini is (in general) objectively _pricier_ than a Toyota. The "value" of a car has more dimensions than simply its associated market price.

All of this is orthogonal to the original issue of $10 being more popular than $100, therefore more valuable.

Except atoms aren't atomic: they can be further subdivided

Those are two different facts:

  1. Atoms exist
  2. Atoms can be divided

> > It means we have a statement for which we can generate additional evidence at will.

Reading the same book over and over again only gives you evidence about what the book says

"Generate additional evidence at will" is packing a lot of stuff, but in my read it means "Generate a hypothesis", "Model a new experiment that can disprove it to the desired or available precision", "Review the experimental setup for any defects", etc etc.

"Generate additional evidence at will" when applied to the "scientific fact that atoms exist" means that any new experiment that you design and perform will be compatible to that fact. There is no evidence whatsoever that atoms (at their "level of abstraction") do not exist.

The only way I see out of that is if you're _so_ reductionist to the point where you accept that the _only_ thing that exists is whatever is at the lowest level of reality (quantum fields, quantum foam, whatever), but then I guess neither you nor I exist to be discussing this.

This is an excellent comment! And this is a *very* strong threshold to reach:

No matter what your threshold for evidence needed is, it can be met.

I'm fearful of mentioning any specific current scientific controversy, but it seems to me that many of the current politically charged topics fail to meet this very strong threshold.

I guess I'm thanking you because while I had an intuitive sense for this, I didn't know this specific wording of "you can meet any threshold for evidence".

My own nonota: https://github.com/matheusd/nonota

CLI TUI, just a list of tasks, track time across them or manually add the time. Backing is a yaml file, so you can adjust if needed.

Runs anywhere you can use Go to compile to.

This is excellent! You should make a T-Shirt with this quote on it!

We could ask! This is HN, which a non-insignificant cohort of various types of investors (angels, VCs, maybe bankers?) visit.

So let's open a couple of Ask HN/Pool HN: Are you an investor? Did you invest in a coop-like startup? Would you invest in a coop-like startup? Under which conditions?