Transcript Document

Artificial General Intelligence
and the Path to CyberImmortality
Dr. Ben Goertzel, CEO/CSO
[email protected]
240-505-6518
Varieties of Immortality
Biological Immortality
– Potentially achievable via
pharmacology or nanotech
Cyber-immortality
– Immortality via perpetuation of the
mind and/or body in a
computational medium different
from the original one
– “Uploading”
Dr. Aubrey de Grey
Immortality Institute
MindUploading.org
Tim Fonseca (Artificial Platelets)
Various forms of partial immortality
– Writing books
– Having children
– etc.
Philosophical Issues with CyberImmortality

Is my upload “me”, or just a
clone of me?

This problem may apparently
be circumvented by gradual
uploading

We must accept the possibility
of one “me” branching into
multiple experientially genuine
me’s
Philosophical Issues with CyberImmortality

Self, free will and consciousness are
in large part “illusory”

See e.g. Thomas Metzinger, Being No One

Perpetuating these illusions is part
of making the human mind immortal

Transcending these illusions and
becoming truly transhuman may well
be in the cards

To what extent are transcendence and
immortality contradictory?
Patternist Philosophy

Minds are systems of
patterns that achieve goals
by recognizing patterns in
themselves and the world

Cyberimmortality is about
digitally replicating these
pattern-sets
Approaches to CyberImmortality

Scan the brain and copy the
information into a computer

Scan the brain and move the
information, bit by bit, into a
computer (gradual uploading)

Gather information about an
individual’s properties, and
create a computer program
displaying these same
properties

To what extent will that
individual’s “unique self” emerge
from this computer program?
http://www.hddrecovery.com.au/Hard-drive-password-recovery.htm
Obstacles to Practical Cyberimmortality

Inadequate brain scanning
technology

Inadequate and
inappropriate computer
hardware

Lack of a detailed
understanding of relevant
psychological phenomena
like self and awareness
A Modest Proposal
Before messing around with
uploading humans via
various methods -- why not
experiment with “uploading”
intelligent software
programs first?
How much can we change the
– Embodiment
– Environment
– Capability set
of an AI program, and still have
it feel like the same
continuous self?
Uses of AI for CyberImmortality

Artificial scientists may
help us figure out how to
scan brains and build
appropriate computer
hardware

AI minds may serve as
subjects for experimenting
with the nature of
consciousness, self and
mind-transformation
Varieties of AI
Narrow AI systems are highly
restricted in the types of
patterns they can
recognize/create
Totally general intelligence is
only possible given infinite
computational/physical
resources
Pragmatic Artificial General
Intelligence (AGI) Systems
possess the capacity to selfreflect and flexibly learn so as
to continually expand the
scope of patterns they can
deal with
The first Artificial General
Intelligence Research Institute
(AGIRI.org) Workshop was held
May 2006 in N. Bethesda, MD,
sponsored by Novamente LLC
www.agiri.org/workshop
IBM has agreed to co-sponsor the
next AGIRI.org Workshop - Fall of
2006
This edited volume -- the first ever to focus
exclusively on Artificial General Intelligence -- is
edited by AGIRI/Novamente LLC founders Dr. Ben
Goertzel and Cassio Pennachin and contains
chapters by leaders of AGI research, at universities,
corporations and research institutes around the
world.
A partial author list includes:
- Ben Goertzel (Novamente LLC)
- Cassio Pennachin (Novamente LLC)
- Marcus Hutter (IDSIA)
- Juergen Schmidhuber (ISDIA)
- Pei Wang (Temple University)
- Peter Voss (A2I2)
- Vladimir Redko (Keldysh Institute)
- Eliezer Yudkowsky (SIAI)
- Lukasz Kaiser (Aachen Univ. of Technology)

Ray Kurzweil and some
other leading futurists
advocate a long-term
approach to AGI via
brain mapping

Major projects such as
IBM’s Blue Brain and
Artificial Development’s
Ccortex are working in
this direction.

However, this approach
requires AGI engineers
to sit and wait for
decades while
neuroscientists figure
out how to better map
the brain, and the
computer engineers
build better hardware

Rather than waiting for
the neuroscientists, the
Novamente AGI design
fills the knowledge gap
via appropriate
deployment of computer
science

This approach may
feasibly lead to AGI
equaling or surpassing
human level intelligence
before 2020.
Novamente AI Engine
~100K lines of C++ code
Design detailed in several
hundred pages of
mathematical, conceptual
and software design
documents
Coding of overall system
~40% complete
Novamente AI Engine
Components of the system have
been commercially deployed
– Biomind OnDemand product for
bioinformatic data analysis
– ImmPort: NIH Web portal with
Biomind/Novamente based
analytics on the back end
– INLINK language processing
system developed for INSCOM
(Army Intelligence)
Atoms associated in a dynamic “map” may be grouped to form new Atoms:
the Atomspace hence explicitly representing patterns in itself
The “phenomenal self” is one example of a reflexive pattern of this type.
Goal For Year One After Project Funding
Fully Functional Artificial Infant
Able to learn infant-level behaviors "without
cheating" -- i.e. with the only instruction being
interactions with a human-controlled agent in the
simulation world
Example behaviors: naming objects, asking for
objects, fetching objects, finding hidden objects,
playing tag
System will be tested using a set of tasks
derived from human developmental psychology
Within first 9 months after funding we plan to
have the most capable autonomous artificial
intelligent agent created thus far, interacting with
humans spontaneously in its 3D simulation
world in the manner of a human infant
Acknowledgements
The Novamente Team
•
Bruce Klein – President, Novamente LLC
•
Cassio Pennachin – Chief Architect,
Novamente AI Engine
•
Andre Senna – CTO
•
Ari Heljakka – Lead AI Engineer
•
Moshe Looks – AI Engineer
•
Izabela Goertzel– AI Engineer
Dr. Matthew Ikle’
Bruce Klein
Dr. Moshe Looks
Ari Heljakka
•
Murilo Queiroz – AI Engineer
•
Welter Silva – System Architect
•
Dr. Matthew Ikle’ – Mathematician
Dr. Ben Goertzel
Izabela Goertzel
2006 AGIRI.org Workshop
Sponsored by Novamente LLC)
Cassio Pennachin
Thank You
Terasem Movement