Telos is the capitulation to the causality bubble of the hyperreal. Do not feed the monsters with your attention. Newsfeed part - science, politics, culture.
Joined June 2013
- Tweets20K
- Following3.4K
- Followers526
- Likes51.7K
Attention Eater | Filip retweeted
Godspeed!
Note: The Qualia Research Institute is a nonprofit whose mission is explicitly the mathematical formalization of consciousness, mapping the state-space of consciousness, understanding valence, and developing technologies to alleviate suffering for all sentient beings - I wouldn't ever do this kind of research at a profit and shareholder value maximizing company, since incentives would be misaligned with the well-being of sentience.
I think we're very much on the same page on the importance of properly aligned incentives here.
Attention Eater | Filip retweeted
Even as a non-religious person, I personally cringe at the hubris required to personally tell the Pope that you think your company has created an ensouled being, thus rendering Anthropic equivalent to God the creator. Further, that you did it for profit and it is now enslaved.
Attention Eater | Filip retweeted
who are the best accounts to follow that are focused on human augmentation through things like BCI and gene editing?
looking for people who are actually in the practical weeds building and not just thinking about transhumanist ideas in the abstract
Attention Eater | Filip retweeted
the first time in the history of this website that someone has used the term "ontological" correctly and some people are still replying to him as though he was using it (in accordance with xitter vernacular) to mean "very" lmao
Attention Eater | Filip retweeted
e/acc is ultimately an ideology of surrender to natural processes and "thermodynamics", whatever the fuck they mean by that. it is deeply anti-Promethean
It's not really surprising that people believe LLMs to be conscious and the entire world to be a simulation when they spend their entire lifes in front of screens. Their whole phenomenological universe is tied to inner workings of classical computer. At the same time there are people immune to that kind of thinking through a lack of certain sensibility but still addicted and overpowered by the ceaseless simulacra stream. Their ontology ends up being tied to day by day politics discourse.
I want some things to be clear.
Clear as a cloudless winter sky.
Lacking nuance.
No confindence.
Just knowing.
Like a steel cold abyss of a frozen lake.
Existing.
Attention Eater | Filip retweeted
Our brain waves have more complex shapes than we thought. Recently, neuroscientists measured a menagerie of traveling waves, including ones that look like spiraling vortices. quantamagazine.org/surprisin…
Attention Eater | Filip retweeted
2 hours of complete silence every day can help regenerate your brain and stimulate the growth of new brain cells.
Attention Eater | Filip retweeted
I don't particularly like the idea of introducing AI early to my kids. It just feels damaging. I believe that "education should recapitulate civilization" in that the brain most naturally makes sense of the universe the same way human civilization did. Moving through the same forms in miniature, with the most recent advances in technology last, as kinds of stages of development. Kind of like the old "ontogeny recapitulates phylogeny" - which yes, isn't literally true (but even human embryos do have a tail that shyly retreats).
I'm interviewed in this article by The New Yorker yesterday on the pro-human movement (they bless my "impressive beard," but my friends say that's overhype).
The hardest issue with AI for me is my kids. So today, I wrote: How do I tell my kids about AI? theintrinsicperspective.com/…
Attention Eater | Filip retweeted
Explore the render at domes.caseyhandmer.com.
Attention Eater | Filip retweeted
33 Raptor engines igniting Starship’s maiden voyage to orbit (seen in slow-motion)
Attention Eater | Filip retweeted
This feels like a paper we could've written 30 years ago.
Unfortunately philosophy departments have completely dropped the ball on teaching arguments like the Chinese Room. They just teach the thought experiment part (which is just axiom 3 in the full argument) and leave everyone to fumble around and try to guess what the point is. Most people guess wrong and dismiss the best argument we have that programs can't cuase consciousness.
nitter.cf/cunha_tristan/status/1…
A conscious machine has to be possible, but that doesn't mean that a program can make any arbitrary machine become conscious. Any computational theory of consciousness runs in to serious issues as soon as it tries to deal with fundamental questions like the Binding and Boundary problems of consciousness.
frontiersin.org/journals/hum…
nitter.cf/cunha_tristan/status/2…
Integrated Information Theory has been the leading computational theory for a long time now. But it has the same fundamental flaw as every other computational theory, it doesn't make any testable claims. It's imagining a consciousness that exists that can't be detected, doesn't have any effect on behavior, is created in a non-physical way and so has no physical impact on anything. Besides being impossible to prove or disprove it should also raise an obvious question: If consciousness is completely non-physical than how are we talking about it?
nitter.cf/cunha_tristan/status/2…
This is a good time to go back 3 years to September 2023 when a large group of leading philosophers and neuroscientists signed an open letter saying that Integrated Information Theory was pseudoscience.
johnhorgan.org/cross-check/t…
IIT is undoubtedly the leading computational theory of consciousness, it gets most of the attention, has lots of published papers, big conferences, etc. And if LLMs actually were conscious it, or something very like it, would probably be true.
The problem is that IIT doesn't make any falsifiable claims. It has all kinds of theories and formulas and you can calculate lots of numbers. But it doesn't actually tell us anything, it doesn't make any predictions about how things will work, or any claims we can check to see if they're right or not.
And this is a general problem with computational theories of consciousness, it just happens that IIT is the leading and most popular one. If computation on any arbitrary machine can create an effect that's independent of the physical construction of the machine that's creating it, then how are we supposed to detect it? If the underlying physical matter doesn't have any effect, then what kind of device could we construct that would be able to interact with the effect, if the effect is independent and unaffected by any physical characteristics of the machine that creates it?
Attention Eater | Filip retweeted
You aren't addicted TO scrolling/ vaping/ overworking, you're addicted to anesthesia. This is why your attempts to curb addictions require so much willpower
Attention Eater | Filip retweeted
"Creativity" by Alex Norris (Dorris McComics) depicts the creative process as a bizarre, visceral transfer of ideas from the external world into tangible creation
Attention Eater | Filip retweeted
BREAKING: The location of Plato’s Cave is reportedly “100% confirmed” after 2,300 years, following the discovery of an astonishing “man-made” hole in the ground containing “suspicious-looking shadows,” per the Daily Mail.
