for Cultivating Benevolent Seed Artificial Intelligences Mark R Waser Digital Wisdom Group Richmond AI amp Blockchain MarkWaserRABCSolutions Agenda Consciousness What is it What is the minimal definition for our purposes ID: 780684
Download The PPT/PDF document "A Collective Intelligence Research Platf..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
A Collective Intelligence Research PlatformforCultivating Benevolent“Seed” Artificial Intelligences
Mark R. Waser
Digital Wisdom Group
Richmond AI & Blockchain
Mark.Waser@RABC.Solutions
Slide2AgendaConsciousness
What is it? What is the minimal definition for our purposes?
How to measure it
Conscious AIWhat we want (and don’t want)How to create it
2
Slide3Where to Start? Do I Invite *You* to Consider
Morality
Self
grow community by developing *others*
an
autopoietic system
Greek, α
ὐτo
-
(auto-)
, meaning "self“ &ποίησις (poiesis), meaning "creation" “A Strange Loop”
3
suppress or regulate selfishness and
make cooperative social life possible
Slide4Inflammatory Statements
I
feel
like the king’s fool – tasked, upon pain of death, to explain the coherent elephant of self, consciousness and
morality
to blind men
The good news is that these are all relatively simple, universal concepts at heart
Admittedly tremendously context-dependent in content thoughThe bad news is that the blind men incorrectly assume themselves to be experts in consciousness and moralityWorse, evolution has programmed the blind men to blindly disagree with any novel views on consciousness and morality (for efficiency & to prevent gaming)I feel, I am small and afraid4
Slide5Disclaimers (Mercier & Sperber)
I am not dogmatic about any of these points.
I do have a goal/agenda. I want to make progress. These facts drive my definitions, my methods and my assertions – but I do try to stay within the envelope of consensus.
My answers are explicitly meant to be one of potentially many – a possible scenario that I would like to share with you. My answers are NOT the only possible answersbut I will be calling out some bad answers & leaky abstractions that are blocking progress.
This presentation was overhauled last night. Any richness and diversity is yours.
My goal is to explicate how consciousness, self and morality combine to create one integral self-reinforcing whole (an attractor in state space) but too little time.
I am claiming that what I’ll present is the capstone of a
solid extant defined archI am small and fallible (and biased), but I believe that *together* . . . .5
Slide6Consciousness is Expensive
Conscious actions are much slower than automatic actions
Conscious actions are much more error-prone and imprecise
Consciousness must have a function/purpose
Consciousness is necessary for “Rational Anomaly Handling”
Consciousness is necessary for learning
RAH & learning == “automated intelligence acquisition”
Temporarily being a “p-zombie” is possible/common but . . .always being a p-zombie is “evolutionarily inadvisable”6
Slide7What it *Does* vs What it *Is*
Functional Description (IIT)
How well does it perform a specific task/operation?
Is *IT* even there or does something else do its job?
Property and/or Content Description (GWT)
How closely does it resemble the archetype of its class?
Its methods are as important as its function
AST is Orthogonalwe think ^we think ^7
Slide8Types of Consciousness
Functional Consciousness
First-Person Consciousness
FunctionalPhenomenalEpiphenomenalInterrupt Consciousness (vs Access Consciousness)Self-Awareness
Intentional Self-Transformation
8
Slide9Morality
Moral systems are interlocking sets of
values, virtues, norms, practices, identities, institutions, technologies, and evolved psychological mechanisms
that work together to
suppress or regulate selfishness and
make cooperative social life possible
9
Slide10Political Moralities (World “Truths”)
Conservative
Liberal
Libertarian
People are
flawed &selfish
inherently good
farseeing and matureMorality isin-group & rule of lawcare for othersfreedomGov’t should besmall & controllinglarge & helpfulminimum necessaryConcerned withshort-termmedium termultimate stateVision of AIOthers see them asselfishfoolishutopian ==> foolish
Philosophy isdeontologicalconsequentialismaxiologyAmygdala or ACC
amygdalaanterior cingulate???
10
Slide11Instrumental Goalssuicide (& abortion?)
masochism
------------------------------------------------selfishness
(pride, vanity)
-------------------------------------------------
acedia (sloth/despair)
insanitywire-heading (lust)wastefulness (gluttony, sloth)murder
(& abortion?)cruelty/sadism-------------------------------------------------ostracism, banishment
& slavery (wrath, envy)----------------------------------------------------slavery
manipulationlying/fraud (swear falsely/false witness)theft (greed, adultery,coveting)
survival
/reproduction
happiness/pleasure
-------------------------------------------------
Community
(ETHICS)
--------------------------------------------------
self-improvement
rationality/integrity
reduce/prevent fraud/counterfeit utility
efficiency (in resource acquisition & use)
and the Eight Deadly Sins
11
Slide12Capabilities Approach (and Questions of Justice)If a country’s Gross Domestic Product increases each year, but so does the percentage of its people deprived of basic education, health care, & other opportunities, is that country really making progress?
If we rely on conventional economic indicators, can we ever grasp how the world’s billions of individuals are really managing?
Have our dominant theories of development given us policies that ignore our most basic human needs for dignity and self-respect?
Are there any better questions than:
What is each person actually able to do/to be?
What real opportunities are available to them?
12
Slide13Qualia & Phenomenal Consciousness
Your entire visual field is an illusion
Appears static & in focus while your fovea saccades
Horribly change blindYour relative times of perception are an illusion
You are prone to false senses of authorship because you can’t directly sense agency
You can’t correctly perceive/retrieve your reasons related to morally-fraught actions
Does pain really “hurt” or does it grab attention and prevent thought/action (which is more important)?
“Strange deficits of consciousness” a la Oliver Sachs are generally perceptual13
Slide14Consciousness and free will are illusions: you are just a robot
If consciousness is an illusion, who is being fooled?
Slide15Phenomenal consciousness is an illusion
ILLUSION
“INCORRECT”
EXPERIENCE
= a trick of perception
== something that most people would accept can happen to a machine
(and clearly happens to humans)
PHENOMENAL CONSCIOUSNESS
15
Slide16Do we see reality as it is?
16
Slide17If Materialism is True . . . .the United States is Probably Conscious
Eric Schwitzgebel
17
Slide18If the United States is Conscious . . . .
It is a tremendously inefficient consciousness
Though, it is efficient in that its focus/attention can be divided
Anomaly-handling and learning really aren’t done at the national levelOther consciousnesses aren’t merely embedded but necessary
Where is there evidence of a unitary phenomenological “sensation” of “self”
It seems as if all sensations (like patriotism) are “felt”/experienced at the individual level
There is absolutely no sign of any “
Unity of Consciousness”SELF^18
Slide19If I Am Conscious . . . .
It is a tremendously inefficient consciousness
Can the national consciousness *really* be divided?
Anomaly-handling and learningMy “society of mind” isn’t merely embedded but necessary
Where is there evidence of a unitary phenomenological “sensation” of “self”
How are my declarations any different from “national pride” or “national sorrow”?
There are only *disputed* reports/narratives of “
Unity of Consciousness”I am only a self-referential process running on my brain cells*Why* isn’t a government running on humans conscious?19
Slide20Designing Consciousness
The United States behaves “as if” it is conscious
We behave “as if” we are conscious
We *believe* that we are consciousWithout any real consensus definition of what that is
Is that belief the only true difference that “makes” us conscious?
What happens when we realize that the United States *is* actually conscious?
What if we design an entity to believe that it is conscious?
20
Slide21Intelligence vs Consciousness
Artificial General Intelligence (AGI)
Can be a tool or an entity
Frame problem & “no free lunch” theorem
The Ultimate is AIXI
Static
& unbounded in time/storage
Goals (& benevolence) are orthogonalDoesn’t innately “care” about anything, derived intentionalityArtificial Conscious Entity (ACE)An autopoietic entity, a “strange loop”, an identityThe ultimate is some sort of self-improving sentient sapienceIt *cares* about itself, it is truly intentional21
Slide22Measurable Functions of Consciousness
Not Just Homeostasis but
Purposeful Self-Modification
to fulfill its identity,
Preparation &
Uncertainty Avoidance
22
Slide23“Classic AGI”
Decisions
Values
Goal(s)
Goal(s) are the purpose(s) of existence
Values are defined solely by what furthers the goal(s)
Decisions are made solely according to what furthers the goal(s)
BUT
goals can
easily
be over-optimized
23
Slide24Decisions
Goals
Values (Identity)
Values define
who you are,
for your life
Goals you set
for short or long periods of time
Decisions you make every day of your life
Entities don’t have singular life goals
24
Slide25AgendaConsciousness
What is it? What is the minimal definition for our purposes?
How to measure it
Conscious AIWhat we want (and don’t want)How to create it
25
Slide26Plutchik’s Psycho-EvolutionaryModel of Emotions
26
Slide27The Digital Wisdom Institute is a non-profit think tank focused on the promise and challenges of ethics,
artificial intelligence & advanced computing solutions
.
We believe that
the development of ethics and artificial intelligence
and
equal co-existence with ethical machines ishumanity's best hopehttps://wisdom.digital27