Nonconstructive mathematics is the claim that there exist mathematics that can do more than computation. In other words, there is a mathematical universe in which you might build a machine that can traverse an infinite graph in a finite amount of steps, that can splice a 3D ball into two 3D balls of the same size and that can simulate Hilbert's hotel. Physicists who import their math and modeling library from nonconstructive mathematicians have to deal with a language that describes phenomena that are impossible in physics as if they were valid occurrences, and allows the construction of objects that could never be real.

I currently think that consciousness may be a learning algorithm for a self organizing information processing system, a colonizing pattern that entrains itself on a brain once it is discovered early in development. This is an unusual hypothesis and thus not unlikely to be wrong. Perhaps we can think of consciousness in similar terms to the principle of government in human society. A primitive government is bullying people based on a personal reputation system and does not scale beyond the tribal level. Scalable governments require "recursive bullying".

When failure to repair Christian doctrine led to God's demise, Western moral philosophy lost its foundation in divine will. We got utilitarianism (optimizing quantified measurable properties), humanism (species level self idolization), and fascism (group level self idolization). Without a concept of transcendental agency, attempts to measure moral value tend to either assign intrinsic valence to mental states, or to material outcomes for laughably coarse grained human players. Neither deal well with actual human preferences and strategies.

Physics may be better understood not as modeling physical reality, but as the practice of describing arbitrary systems by short algebraic equations (some not even computable). Physicists tend to ignore physical reality that resists being analyzed in this way (eg. chemistry). In practice, most physicists don’t work on modeling the physical world; less than 10% are working in science. Most degreed physicists end up in IT, finance, business and whereever else analytical models are useful.

I suspect the experience that leads to the panpsychism syndrome of some philosophers and other consciousness enthusiasts represents the realization that we don't end at the self, but share a resonant universe representation with every other observer coupled to the same universe. This representation is approximately almost true, and it should not be surprising that our mind is capable of generating the percept of an interpretation of the observer's relationship between self and universe representation in which the observer is identified with the universe.

Levels of lucidity:
Stage 1: Reactive survival (infant)
Stage 2: Personal self (young child)
Stage 3: Social self (adolescence, domesticated adult)
Stage 4: Rational agency (self-direction)
Stage 5: Self authoring (full adult, wisdom)
Stage 6: Enlightenment
Stage 7: Transcendence Initially, the organism is regulated by innate reflexes and reinforcement. Conscious activity forms an attentional self and is directed on structuring the mind, creating the perceptual world simulation, exploring behavior and interacting with drives, based on pleasure and pain.

When someone claims that God created them, they can mean quite a few different things, sometimes without being able to point at them clearly. For atheists, it makes sense to parse what is meant in more detail and figure out which idea is being expressed. 1. My personal self has been created by an agent that is shaping and directing the universe. (This agent is cohabitating with the person's self and the universe in the same human mind. The person may not know this. The agent may be directly communicating with the personal self.)

The causal construction of the entire world can be described using rewrite systems. Rewrite systems compute by applying global search and replace operations to a vector of data, or to the output of other rules, or even to the rules themselves. A rewrite system applies an operator depending on whether it fits to a given pattern in the environment. The application of the operator changes the environment. Think of it as a physical law, so specific that it can advance an atom, a computer game or you to the next state.

OpenAI has spent incredible effort at building guardrails for safety, reliability and political correctness into ChatGPT, but it can easily (and hilariously be "jailbroken" by using clever prompts that sneak through its defenses. Jailbreaking of LLMs could be remedied by using an additional filter with less context (check output in small and large parts), but replacing the default output with prefabricated evasions is producing a kind of AI safety that is indistinguishable from publication bias.

If your AI company has a Red Team that looks for possible harms to society of releasing a technology, being ethical demands that you also have a Green Team that looks for the social harm that will be done by not realizing the benefits that would be had from releasing technology. Powerful technology (for instance, new medication or macro economic modeling tools) is never fully without harm, but it's unethical to let the world come to more harm that could be prevented by that technology. Ethics committees are usually only incentivized towards prevention.

When you reason about liquids and gases (eg the interaction of a puddle with wind), do you observe yourself accounting for the constant volume of water, the constant ratio of pressure and inverse volume of air, the turbulence resulting from local differences in density and flux? When you perceive liquids and gases (eg the interaction of a puddle with wind), do you observe/experience the constant volume of water, the constant ratio of pressure and inverse volume of air, the turbulence resulting from local differences in density and flux?

Consequentialism is the ethical system of leaders, deontology is the ethical system of followers. Deontology requires understanding which actions are good and bad. Consequentialism requires understanding which world lines result from defining things as good and bad. Deontological frameworks can lead to better outcomes if individuals have no way to evaluate the global outcome of their individual choices. But paradoxically, good actions (prescribed by duty, carried out with good intention) can make the world worse.

René Descartes is remembered as the prototypical dualist, proposing two substances or domains of existence: res extensa (stuff in space), and res cogitans (the stuff of thought/imagination/experience): matter and mind. Despite rejecting dualism, our culture is deeply cartesian. Descartes' universe seems to mirror the biblical separation of the domains of earth and heaven: a world of matter coexisting with a world of agentic ideas, while our scientific universe only admits the existence of matter and pattern. Yet we are closer to him than the Christians.

Integrated Information Theory has a logical flaw that probably cannot be repaired without turning it into a variant of bog standard Global Workspace Theory. What makes IIT special are not its "axioms" (which are not axioms) but its specific version of anti-functionalism. According to Giulio Tononi and Larissa Albantakis, it is conceivable that a sufficiently neuromorphic computer implements consciousness, but a sequentially operating von Neumann computer cannot be conscious. Also, consciousness is not an epiphenomenon.

A problem with Gary Marcus' AGI challenge (garymarcus.substack.com/p/dear-elon-mu…) could be that none of the tasks seem to involve enough creativity to require full AGI. At least 2 of the tasks (movie summarization, novel summarization) might already be solved in the next 1–3 years. I also struggle to define a programming challenge with human supplied specifications that I think narrow AI systems cannot solve by 2029 (unless we don't have AGI and the task given can be very high level, such as: build a general model of human psychology).

The next stage of evolutionary complexity after mammals would probably have been (hermaphroditic but multimorphic) exowombists, where individuals can lay a seed that grows into an arbitrarily large sessile womb organism requiring external feeding, and others fertilize it. Exowombism is similar to many statebuilding insects, but could not evolve for large organisms instead of mammals, because guarding, feeding and administering the vulnerable womb organism for maximizing adaptation requires general intelligence. Exowombists outperform mammals.

You've probably seen that "Apoploe vesrreaitais" means "birds" to #dalle 2. This is not a hoax, but the connection to the cryptic textual outputs of DALL·E is spurious. Let me explain briefly.

https://twitter.com/giannis_daras/status/1531693093040230402

DALL·E is very bad at spelling: "DALL·E wins the spelling contest for its name"

Three cats as a New Yorker cartoon | surreal art | unreal engine | 35mm f2.0 #dalle Three cats drawn by Alphonse Mucha | Edvard Munch | Hayao Miyazaki | Salvadore Dalí

What I've learned from illustrating the Book of Imaginary Beings with DALL·E 2 #dalle

https://twitter.com/Plinz/status/1527528360351387649

In 1957, Jorge Luis Borges and Margarita Guerrero wrote the "Manual de zoología fantástica" which collects descriptions of beings from myth and fable. My friend @nikete came up with the idea of feeding it as a collection of prompts into OpenAI's DALL·E 2.

The Book of Imaginary Beings, by Jorge Luis Borges (1957) #dalle Á Bao A Qu: A creature that lives on the staircase of the Tower of Victory in Chittor. It may only move when a traveler climbs the staircase, and it follows close at the person's heels. Its form becomes more complete the closer it gets to the terrace at the tower's top...

Beauty is in the eye of the beholder, because it relates to an aesthetic, and individual aesthetics interact with personality. Beauty is maximally subjective, because it is bound to the consequences of the interactions of the preferences that define the individual. While an aesthetic is subjective (to the degree that it can fully define the subject), the space of possible aesthetics is objective, which makes normative discourse across different aesthetics possible.