Certhas

20 ms·

1.
by Certhas 5d ago
Anthropomorphizing is problematic because a human mind is a very bad model for what LLMs are. A lawnmower is a much much much worse model.
2.
by Certhas 6d ago
I have never come across this position argued seriously. I would also consider it absurd, as then the question whether we have consciousness depends on unknown properties of fundamental physics (which is not incompatible with determinism, s
3.
by Certhas 6d ago
I am a physicist. I worked (briefly) on foundations of quantum mechanics. I discussed compatibilism extensively with philosophers. I have _never_ come across the position you seem to take here, that determinism has bearing on the question i
4.
by Certhas 6d ago
Nonsense. There was one proposal for relevant quantum effects in brain dynamics, and that turned out to be not relevant. Even if they were, you could substitute all quantum randomness with pseudo randomness and obtain an absolutely indistin
5.
by Certhas 7d ago
Your car analogy is very, very confused. If a simulation of a car can get you from A to B, requires the same steering, fuel and servicing, gives you the same tactile feedback, is it a car? Of course "I care about humans because I am hu
6.
by Certhas 7d ago
Videos and LLMs are not deterministic in the same sense at all. LLMs are deterministic in the same sense as biological processes. And a faithful simulation of a brain would have all the properties you note.
7.
by Certhas 7d ago
I generally agree about the problem with anthropomorphizing. But I don't think Anthropic are doing that. They explicitly write "in biological entities this would be considered a sign of consciousness, but we don't know how to
8.
by Certhas 7d ago
I like it, and it points in the right direction, but is not directly true: The markers are about interactions, how biological organisms behave in certain test situations. But it speaks to the central question: Are the tests adequate? Or are
9.
by Certhas 7d ago
Humans are biological machines that generate further humans. Lawyers and diplomats and politicians and bureaucrats are humans, that only generate text. We are seeing LLMs have cognitive abilities that significantly exceed human abilities. A
10.
by Certhas 7d ago
I believe it's deeply serious, and the scientifically correct stance. Especially the observation: "Claude exhibits markers in its behaviors, self-reports, and internal representations that we would consider welfare-relevant if obs
11.
by Certhas 7d ago
What's your definition of sentient? Or, maybe more precisely, consciousness? I think it's reasonable to at least start thinking about these questions. It has long been established that LLMs have good theory of mind [1]. And there
12.
by Certhas 8d ago
Please explain how this has nothing to do with LLMs and Computers?
13.
by Certhas 9d ago
If I am reading your question correctly you are asking about chat interface Vs Codex/Claude code? If so, in my experience Codex/Claude code use is widespread for mathematicians who are seriously using these tools.
14.
by Certhas 9d ago
This is OpenAI though. There were zero visible consequences to them unleashing a swarm of agents on the public internet. We will see how it goes down but my prior is zero consequence and a statement along the lines of "Isn't our A
15.
by Certhas 9d ago
The point of the local model in the context of the article was to argue that you can't ban these capabilities. Making datacenters and public clouds only rent GPUs to a restricted list of people, while tightly monitoring what people do
16.
by Certhas 13d ago
This is such an absurd take given what we know about the hugging face attack. The problem has emphatically not been that someone was misusing the technology.
17.
by Certhas 13d ago
I spent/spend a lot of time in the UK. I already agreed (as did the UK Government audit) that there was a reluctance to investigate due to fear of being racist. But the official inquiries have happened, and they were initiated by the l
18.
by Certhas 14d ago
In Germany we have a term "besorgte Bürger" - concerned citizens. They just share their concerns. Concerns about all the brown people. Immigrants. Crime. Just sharing concerns. But the blog posts by DHH that are being linked here
19.
by Certhas 18d ago
Feels like mostly LLM work...
20.
by Certhas 22d ago
Imagine Anthropic gives you Opus of 6 months ago but at much higher speeds and much lower cost (that they might or might not pass on). Would you use it?
21.
by Certhas 22d ago
It's not? Schwarzschild has a space-like singularity. That's the wiggly horizontal line at the top left of the diagram. If you are in the black hole you can't avoid hitting it. Seems to be exactly what the paper is remarking
22.
by Certhas 22d ago
As is nicely visualised by it's Penrose Diagram, e.g. https://jila.colorado.edu/~ajsh/insidebh/penrose_schw.gif
23.
by Certhas 28d ago
I didn't realize that, thanks! After looking at Wikipedia for a bit it seems there are two phases under the PCT, an international phase that is sort of like the main part of a patent application, and then the national phase that actual
24.
by Certhas 29d ago
The whole EU censorship rhetoric is almost always devoid of facts and examples. I feel like it is mainly sustained because US people like the narrative that theytl are the free(TM).
25.
by Certhas 29d ago
Presumably it's an "international patent" filed in Brazil. You don't have to file your patent in every jurisdiction. There have been treaties for recognising each others intellectual property rights since the late 19th c
26.
by Certhas 1mo ago
Posts like this truly boggle the mind... The US is in a class of its own when it comes to health spending. The second highest OECD country (per GDP) is Germany which is 5 points lower than the US (and BTW these figures include public and pr
27.
by Certhas 1mo ago
The US spends more than 16% of GDP on health. The UK around 11%. Shrinking US health expenditure by 3% of GDP while roughly maintaining health outcomes is eminently, obviously doable.
28.
by Certhas 1mo ago
I don't doubt that ISO models are not optimal, but the annual cost of power outages to US consumers is above 100 billion dollars last time I looked already. Electric system reliability is a place where you have to be somewhat conservat
29.
by Certhas 1mo ago
Elon Musk has actively endorsed the right wing extremist party in Germany. The one that often has trouble running it's events as businesses try to avoid serving them. And which has prominent members that the secret service considers de
30.
by Certhas 1mo ago
The design and intellectual heritage of Julia and Mojo couldn't be more different though. Mojo is a Swift/Rust style language from industry veterans, Julia is a Lisp built to make academics write faster code.

More ›