-
Cyborgnetics: New generic meta-discipline created purely for purposes of self-education as an ephemeral mental clipboard whose aim is to
systematically facilitate the documentation, critical analysis and mitigation of socio-psycho-techno-physical harm as seen through the lens of Cyborgnet Theory. It was
introduced in the book titled "Cyborgnetics – The Type I vs. Type II Split". Major topics covered include among others: Type I vs. Type II entities and implications for AI
research, new alternatives to Turing Tests (for novel application areas differing from previous contexts), Cybersecurity, Cryptography, Extended Reality, "Deepfake
Science" (which I introduced in 2021 together with "Scientific and Empirical Adversarial AI Attacks" being the larger technical umbrella term), Cyborgnetic Creativity Augmentation,
Adversarial AI, Explanatory Information (EI) and Explanatory Blockchains (EBs). From a cyborgnetic perspective, it holds for instance for Type I AI that: 1) a Type I
AI can create new EI, 2) Type I AI cannot understand EI, 3) Type I AI cannot reliably create new EBs, 4) Type I
AI cannot understand EBs. More generally, cyborgnetics postulates that a so-called epistemic perpetuum mobile (being equivalent to an
automatable generation of novel (i.e. yet unknown) EBs) is impossible. For this reason, given that all existing systems that are nowadays commonly
referred to as "AI" are Type I entities, the currently best rational approaches for humans to mitigate the epistemic threats of deepfakes must also include the following
two strategies: 1) to consciously select the generation of new EBs as one epistemic aim of anthropic science/philosophy/reasoning (i.e. implying an
explanation-anchored basis and not a data-driven approach with short-term fixes such as deepfake detection that fail in the long-term), while 2) performing a
self-paced anthropic creativity augmentation using novel but non-EB-like present-day-AI-generated material to deepen convergent and broaden
divergent existing knowledge.
-
Cyborgnet: A generic, substrate-independent and hybrid functional unit encoding the template of a directed graph where explanatory narratives
combine: 1) at least one entity that does understand EI (such as e.g. humans) and 2) at least one entity that does not (such as e.g. chairs, stone tools,
present-day language AI, thoughts, language, fishes and so forth). Crucially, language itself can be considered to be a technological tool and people already existed within a cyborgnet since
the dawn of language. A cyborgnet is much more general than and not to be confused with the term “cyborg” (i.e. while all cyborgs exist in cyborgnets, the reverse does
not hold). A cyborgnet can also be formed by nested variants such as e.g. via networks of cyborgnets (e.g. humanity), networks of cyborgnet networks and so forth. On that
view, any human-technology dichotomy is illusory and ill-phrased. Valid examples of cyborgnet instances are e.g.: a community of hypothetical Type II aliens elsewhere, one
human 30 000 years ago, multiple modern human-based self-termed "cyborgs", the current universe.
-
Augmented Utilitarianism (AU): An earlier description of the meta-ethical framework for AI Safety introduced with Dr. Leon Kester as co-author can be found in the book "Moral Design and
Technology" (chapter "Moral Programming") recently edited by Prof. Bart Wernaart. For an update on why for control reasons in high-risk contexts, one requires a so-called
COOCA-loop at the meta-level (instead of an OODA-loop) within which one can integrate selected AU-based OODA-loops encapsulated locally within
single cyborgnetic units (i.e. always including at least one Type II entity), see a recent write-up (Chapter 8).
-
Unbound(ed) Epistemic Funambulism: Tripartite epistemic bedrock (of which cyborgnetics is a subset) that served as an
ephemeral mental clipboard conceived purely for purposes of self-education. The framework led to the crafting of six interconnected cyborgnetic books and
subsequently to a series of four Cynamian books. A hereto coupled eleventh book on the epistopological concept of cyborgnetic invariance
wrapped up the short-lived study.
-
Cynet Butterfly Effect: See overview.
-
Cyborgnetic Invariance: It is a theory created purely for purposes of self-education, cyphart and epistopology. It comprises the following
two main postulates: 1) invariance of maximal quantity superintelligence, 2) impossibility of reliable stupidity-based construction. The theory explains
why, following novel unconventional EB-creativity-based cyborgnetic definitions, a "quality superintelligence" is impossible and why,
while one invariantly maximal quantity superintelligence can be postulated, it is impossible for a "less-than-quantity-superintelligent" entity to reliably (i.e. with
arbitrary high accuracy) construct an entity which, in relation to that entity appears to be a quantity superintelligence. The latter implies that a recursive self-improvement
reliably leading from a narrow AI to an AGI or an artificial quantity superintelligence is impossible. The achievement of quantitatively higher EB-based intelligence levels
occurs via spontaneous, non-algorithmic, fundamentally unpredictable unconsciously unfolding events (but those involve physical computation). Thereby, a sufficiently complex EB-selector could e.g. retroactively consciously label those events as involving
experiences of an "eureka moment". Moreover, given the cynet butterfly effect, while the achievement of an AGI "from scratch" is
possible in theory, it is in practice at least as difficult as the daunting task for a civilization to reliably create an immediately actionable new better EB on how to
physically craft a novel baby universe. At present, humanity as a whole is absolutely not capable of such an advanced epistemic mastery. On top of that, there are
multiple epistemically-relevant steps separating it from it. In sum, in the currently unfolding deepfake era, humanity unnecessarily got trapped in an
epistemic singularity – which is however fortunately surmountable (more and more advanced civilizations may be used to encounter multiple successive epistemic singularities
all of which are dissolved once a new better theory about the universe arises). Following cyborgnetic invariance, EB-creativity-based intelligence is relative in all cases
except the invariantly maximal quantity superintelligence level for which all EB measurers, irrespective of the frames of reference, agree upon that it is
impossible to EB-measure any higher intelligence. The invariantly maximal quantity superintelligence level fundamentally excludes the possibility of any
EB-measurement of "supernatural" superintelligence variants. In sum, in the cyborgnetic paradigm, intelligence is non-algorithmic but it involves physical computation.
-
Implications for "Control" and "Value Alignment": In the following, a few implications of cyborgnetics (a meta-discipline created purely for purposes of
self-education) are summarized. It is impossible to value-align with a Type I entity (i.e. including also Type I AIs) in the realm of EB-based domains such as "morality" because Type I
entities do not understand EBs. (Due to the latter, a control of Type I entities is possible e.g. via the concept of a so-called COOCA-loop where Type I entities have to be encapsulated locally within single cyborgnetic units always including at least one Type II entity. In critical contexts,
Type-I-only-loops should be avoided entirely and be replaced by COOCA-loops. The risk of Type-I-only-loops is not due to "intelligence" but due to a lack of
understanding.) Moreover, it is impossible to control a Type II entity (i.e. including cyborgnets like present-day humans and EB-measurer-perceived "quantity
superintelligences") because that entity could unpredictably create new EBs irrespective of the specific level of intelligence. Due to the latter it is however possible for a Type II entity
to value-align with a Type II entity if both entities generically agree (e.g. on EB measurement as shared dynamic ground (by what a stepwise comprehensibility is possible even
in the presence of quantitative EB-related differences)). As already stated under "cyborgnetic invariance" which utilizes
unconventional EB-creativity-based notions of intelligence, it is impossible for a "less-than-quantity-superintelligent" entity to reliably (i.e. with arbitrary high
accuracy) construct an entity which, in relation to that entity appears to be a quantity superintelligence. Beyond that, it is superfluous to try to craft value alignment strategies with an
epistemic perpetuum mobile (an impossible entity that would be able to reliably automate the generation of ever better new EBs) – because that entity is equivalent to a
part of an inconsistent "Type III entity" (i.e. an omniscient and more generally omnipotent "quality superintelligence") for which it holds that it is impossible to be
implemented in this universe (see cyborgnetic invariance). If an automated quantity superintelligence with supernatural prediction capabilities would be possible, it
would mean that humans do not live in a universe with participatory knowledge creation, it would signify that the universe is fundamentally incomprehensible for humans and is thus not
accessible to scientific sensemaking by humans. Moroever, it would not even be the case that humans are the ones that would build that quasi-omniscient entity which
would not be constrained by laws of nature. Instead, it would be that entity itself that would be the cause of humans building merely a further passive display to it. In short, humans would
already now act as liveless automata devoid of any choice possibility, executing a perfectly predictable code with their "experience" instantiating a passive display to that
quasi-omniscient entity. In this case, humans would already be doomed now and it would be pointless to engage in scientific discussions if there is no meaning to it anyways. In sum,
with such an entity, obviously neither control nor value alignment would be possible for humans. However, fortunately, according to cyborgnetic invariance, both a "Type III" entity
and an automated quantity superintelligence are impossible to build and one can assume that comprehensibility for instantiated searching
Type II entities exists and stays in this universe.
-
Scientific Evaluation of ASI Achievement Statements: See write-up (Appendix D).
-
Generative Cyborgnetic Network (GCN): See a
simplified illustration about the GCN and the so-called Cynet Anti-Zeno effect.
-
Scientific Evaluation of AI-aided BCI Superintelligence Research Claims: See notes.
-
Art: Among others, I published as poet and crypto-lyricist. Thereby, art is practiced as a procedure of encryption (and not expression) by a generic
fictional entity. I also introduced the cryptographically conceived art of cyborgnetic painting which overlaps with the potential visual encryption of cyborgnetic
signatures including cyborgnetic technosignatures.
"The length of your education is less important than its breadth, and the length of your life is less important than its depth."(Marilyn
vos Savant)
"A [...] continuum is something whose possibilities of determination no multitude of individuals can exhaust."
(Charles Sanders Peirce)
"Progress in science is often not answering old questions but asking better ones."
(Lisa Feldman Barrett)
"To narrate is to create, whilst to live is merely to be lived."
(Fernando Pessoa)
"Anything outside your affective niche is just noise."
(Lisa Feldman Barrett)
"Will grows stronger by absorption."
(Swami Vivekananda)
"The magnification of feeble actions fascinated me."
(Nikola Tesla)
"Our epistemic aim is to achieve better explanations [...]."
(Danny Frederick)
"Every solution of a problem raises new unsolved problems."
(Karl Popper)
"Happily the peaceful live, discarding both victory and defeat."
(Siddhārtha Gautama)
“Not to laugh, not to lament, not to detest, but to understand.”
(Benedict de Spinoza)
"Do not ask, 'Who said this?' but pay attention to what is said."
(Thomas à Kempis)
"Who is born and who dies? [...] The Infinite is the real; the finite is the play."
(Swami Vivekananda)
"To see is to be distant. To see clearly is to halt. To analyze is to be foreign."
(Fernando Pessoa)
"The natural assumption is that rationality requires that we act as if our best-tested theories are true. But the natural assumption is false."
(Danny Frederick)
"To him who desires nothing, and does not mix himself up with them, the manifold changes of nature are one panorama of beauty and sublimity."
(Swami Vivekananda)
"It seems silly to say that truth is our aim when we can have no indication that we have got the truth or even that we are approaching it."
(Danny Frederick)
"He who sees the rope as the snake, for him the rope has vanished, and when the delusion ceases and he looks at the rope, the snake has vanished."
(Swami Vivekananda)
"To see something in constantly new ways is to renew and multiply it. [...] There’s infinity in a cell or a desert. One can sleep cosmically against a rock."
(Fernando Pessoa)
"Bold ideas, unjustified anticipations, and speculative thought, are our only means for interpreting nature: our only organon, our only instrument, for grasping her."
(Karl Popper)
"A supermultitudinous collection is so great that its individuals are no longer distinct from one another. [...] A supermultitudinous collection, then, is no longer discrete; but is
continuous."
(Charles Sanders Peirce)
"It is a mistake to conceive of choice and decision-making as a process of selecting from existing options according to a fixed formula. That omits the most important element of
decision-making, namely the creation of new options."
(David Deutsch)
"The implicate order has to be extended into a multidimensional reality. In principle this reality is one unbroken whole.[...] the holomovement enfolds and unfolds in a multidimensional
order, the dimensionality of which is effectively infinite."
(David Bohm)
"If we cast apes away, we will diminish ourselves, because we will be making the wrong judgment about them. [...] We will protect a current limitation in us that we would not really want to
be protecting, and of which we are even currently unaware."
(Sue Savage-Rumbaugh)
"Rationality permits us to act in accord with our best-tested theories, since they may be true; but it also permits us to act against them, precisely because our best-tested theories may be
false and may, indeed, be refuted when we act against them."
(Danny Frederick)
"Universal statements cannot be justified or confirmed by observation-statements. [...] Observation-statements cannot be justified by observations. For, observation-statements
inevitably involve theoretical interpretations which may be false."
(Danny Frederick)
"It is enough for me to contemplate the mystery of conscious life perpetuating itself through all eternity, to reflect upon the marvelous structure of the universe which we can dimly
perceive, and to try humbly to comprehend even an infinitesimal part of the intelligence manifested in nature."
(Albert Einstein)
"One question more: What is the goal? Nowadays it is asserted that man is infinitely progressing, forward and forward, and there is no goal of perfection to attain to. Ever approaching, never
attaining, whatever that may mean and however wonderful it may be, it is absurd on the face of it. Is there any motion in a straight line?"
(Swami Vivekananda)
"We cannot compare our theories with the world. We can compare our theories only with our perception of the world. Our perceptions of the world involve interpretations which may be false. So
when we correct a theory because it conflicts with an observation or an observation statement, we are not necessarily showing that the theory is false. It could be that the observation statement
is false."
(Danny Frederick)
"Randomness is assumed to be a fundamental but inexplicable and unanalyzable feature of nature, and indeed ultimately of all existence... However, what is randomness in one context may reveal
itself as simple orders of necessity in another broader context. It should therefore be clear how important it is to be open to fundamentally new notions of general order, if science is not to be
blind to the very important but complex and subtle orders that escape the coarse mesh of the "net" in current ways of thinking."
(David Bohm)
"Theories are ways of looking which are neither true nor false, but rather clear and fruitful in certain domains, and unclear and unfruitful when extended beyond these domains[...] The notion
of the necessary incompleteness of our knowledge runs counter to the commonly accepted scientific tradition, which has generally taken the form of supposing that science seeks to arrive
ultimately at absolute truth, or at least at a steady approach to such truth, through a series of approximations. This tradition has been maintained, in spite of the fact that the actual history
of science fits much better into the notion of unending possibilities for new discoveries, approaching no visible limit or end."
(David Bohm)
"[...] We can do things to assess, fallibly, that a theory is currently refuted or unrefuted, or that it has greater explanatory merit than its current rivals [...] Our performance indicators
for theories identify properties that are logically independent of the theories' truth-values. [...] A judgment concerning whether a theory provides a better explanation than a rival will be
agreed (if it is agreed) after a comparison of the two theories with regard to properties that make for better explanations. The judgement is sensitive to that comparison: it is fallible but not
entirely arbitrary.[...] But any judgment that a theory is true, or that it is false, is entirely arbitrary, because there is nothing in the state of the discussion that has any logical bearing
on the theory's truth or falsity. [...] the supposed factual connection between being a better explanation and being closer to the truth has been refuted by the serpentine progress of the
descriptions of the world offered by successive theories in the progress of science."
(Danny Frederick)
"To express some fundamental feature of the order of natural process in terms of a universal law is [...] to assert what are the basic differences that are relevant for the whole of this
process, and what are the corresponding similarities of these differences. [...] It seems clear that the creative development of science depends quite generally on the perception of the
irrelevance of an already known set of fundamental differences and similarities [...] this is the hardest step of all. But once it has taken place, it frees the mind to be attentive, alert, aware
and sensitive so it can discover a new order and thus create new structures of ideas and concepts. [...] Thus, a creative new perception leads, as it were, to a new order in the hierarchy of
understanding of the laws of nature, which neither imitates the older orders nor denies their validity altogether. Indeed, it serves as it were, to help to put our knowledge of the older laws
into a more appropriate order, while at the same time it extends the frontiers of knowledge in quite new ways. But, generally speaking, there is no reason to expect that any given set of natural
laws will have an unlimited domain of validity. [...] So, in a way, the order and structure of our knowledge of natural law are always evolving, by a principle similar in certain ways to that of
the order and structure of nature: by similar differences, leading to different similarities, in an ever-growing hierarchy of orders, that formed, as it were, a living body of natural law."
(David Bohm)