Other Fediverse accounts:

  • 2 Posts
  • 49 Comments
Joined 1 year ago
cake
Cake day: June 10th, 2023

help-circle










  • That relies on human brains that are trained. LLMs are not human brains. “Training” them is not the same thing as teaching humans about something.

    Circular reasoning. “LLMs are different from human brains because they are different”.

    Also, why did you felt compelled to add the adjective “human”? Don’t you consider that gorillas, dolphins, octopuses or dogs are intelligent, capable of learn new things?

    Human brains are way more complicated than just a bunch of weighed correlations.

    And that is the problem of your argument. You seem to believe that intelligence is all-or-nothing, that anything that hasn’t a human-level intelligence is not intelligent at all. Of course human brains are more complicated that current LLMs, nobody has ever disputed that. But concluding that they aren’t and will never be intelligent because they aren’t as complicated is a huge non-sequitur.





  • Things like asteroids, galactic dust and the like are already accounted for in the baryonic (ie ordinary) matter. We can estimate it for example measuring the absorption of different wavelengths of light, or extrapolating the local abundance of asteroids. There are theories like the MACHO that propose that we are missing some, but in general it is understood they can only account for a tiny fraction of the missing mass.

    The predominant hypothesis is that dark matter is composed by some unidentified particles, that have the same thermodynamic properties as usual matter (basically that their energy is proportional to the volume), but that don’t interact (or interact very weakly) with normal matter.



  • What you’re talking about is the energy dependence of the coupling constants, which is a phenomenon that is very well understood theoretically, and also checked in experiments. The early universe was much hotter, and thus particles had much more kinetic energy and “felt” slightly different coupling constants. The neat thing is that, since this is a purely energy-dependent effect, we can recreate the conditions of the early universe: the collisions at LHC have an energy of the order of 1 TeV, which corresponds to a temperature of 1016K, the temperature 10-12 s after the Big Bang. Anything after the first 10-12 s we can directly recreate, and from 10-12 s to about 10-30 s-ish we can more or less reliably extrapolate. And of course this is all included in the standard Lambda-CMD cosmology.

    Although the article is behind a paywall (which is somewhat strange in cosmology, but I digress), you can check other articles by the same author that also use the “varying constants” framework, for example https://arxiv.org/abs/2201.11667. His framework is that the speed of light c, the Planck constant h, the Boltzmann constant k and the Gravitational constant G depend directly on time, or to be more precise, on the expansion factor of the universe. There are two big differences with respect to what you were saying:

    • c, h and k are not coupling constants, and therefore they don’t receive any energy-dependent corrections. In fact, you could think of these constants as “conversion factors” between units: c converts space-time coordinates in seconds to space-time coordinates in meters, k converts kinetic energy in Joules (or electronvolts) to kinetic energy in Kelvin, and h converts angular momentum or action measured in quanta to angular momentum or action measured in J·s (or eV·s). Honestly it doesn’t make much sense to me that these constants could change (what does it means, in physical terms), that they could change in a correlated way, or that they could change in a correlated way to one, and only one, coupling constant, G.
    • Since this is a time-dependent change, there is no real way to significantly test the hypothesis (unlike the energy-dependent changes). We can not go back in time, or to wait to a different time when these constants would be different. He actually proposes to study how the experimental determinations of these constants in the last 10-20 years, which sounds very wild, as those tiny differences are very susceptible, by definition, to experimental uncertainties, and they are not very suitable for controlled tests.

  • Yes, this is a bit frustrating. Part of the scientific method is to propose new hypothesis, and that’s what the original author did, so no issue with that. But then there is a chain of increasingly pop-sci media that hype some of these hypotheses as they were already confirmed and accepted by the mainstream scientific community, which is not the case. For example, the title of the article, “New research puts age of universe”, that is pure clickbait, the correct tense is, being very generous, “could put”.

    And when this happens in a field like cosmology, it’s relatively harmless. But the same happens in fields that have a more direct impact in the general public’s life, like the usual “a couple of years ago they said eggs were unhealthy, and now they say we should eat 5 per day”. And the effect is that people stop trusting the recommendations of the experts.