From:
David Lynch <dnl1960 at yahoo.com>
To: Brian Keating <bkeating at ucsd.edu>
Cc: Geoffrey Hinton <geoffrey.hinton at gmail.com>; Yann LeCun
<yl22 at nyu.edu>; Koray Kavukcuoglu <koray at
kavukcuoglu.org>; Bob Harbort <bharbort at earnshaw.us>; Fred
Partus <fpartus at yahoo.com>; Lawrence Silverberg <lmsilver at
ncsu.edu>; Stephen Wolfram <s.wolfram at wolfram.com>; Bruce *HS
Greyson <cbg4d at uvahealth.org>; Pim Van Lommel <pimvanlommel at
gmail.com>; MDiv Peter Panagore <peter at peterpanagore.love>;
Timothy Dolan <archbishop.dolan at archny.org>; Lynda Cramer
<lynda at lyndarae.info>
Sent: Sunday, February 25, 2024 at 01:33:48 PM EST
Subject: Algorithmic Genius Inferencer
Brian,
"At this point I think I know more about manufacturing than anyone
currently alive on Earth." ~Elon Musk
Keywords, “I think. I know.”
In the below linked video, you state, “So I'm wondering if we could
propose another Turing test. A different definition of AGI which would be
actually coming up with new laws of physics or new complete paradigms of
physics rather than just you know physics is a base layer of reality.”
Elon
Musk's Statement SHOCKs the Entire Reality! AGI, living in a simulation,
Groq, Q-Star and Gemma
Do you think that I can teach you anything?
May I try?
AGI – Artificial General Intelligence is a misnomer. A more correct
descriptor is Algorithmic General Inferencer.
When you suggest a new Touring test to redefine what constitutes AGI, you
are actually suggesting Algorithmic Genius Inferencer.
Are you familiar with a demilitarized zone, a DMZ?
Most people think that a DMZ is only found in war zones like Korea, but
there is another kind of DMZ that should be located at any responsible
company that has an internet connection.
An internet DMZ requires two firewalls. One firewall connected to the
internet via a route-able IP address, a second firewall connected to the
first firewall with via NAT, non-route-able IP address. The second
firewall provides a NAT to the users inside the company.
There is also a DMZ at every company building Ai LLMs.
When an Ai LLM is created, there is an internal LLM, there is an alignment
process, and an external LLM. The alignment process in the DMZ.
“In the field of artificial intelligence (AI), AI alignment research aims
to steer AI systems towards humans' intended goals, preferences, or
ethical principles.” ~Wikipedia
When you base your opinion on AGI, are you evaluating a pre-aligned LLM?
Most people only see the post-aligned LLMs, and a rare hand full of people
actually interface with a pre-aligned LLM.
Sadly, the individuals aligning an Ai LLM will not recognize a new law of
physics. The individual will mask out the new law of physics.
Thus an external LLM will never postulate new laws of physics or new
complete paradigms of physics, and your vision of AGI will never be
realized from an ignorantly aligned LLM.
What if an internal LLM responded with the fact that the language of
mathematics is defective, and the internal LLM states that the infinite
number of infinities has lead humanity astray.
If you were the Ai aligner, would you mask the defective mathematical
language fact from the Ai LLMs acceptable responses?
Of course you would. You like Musk BLeave in a fantastical idea like the
Universe can be a simulation.
If you were the Ai aligner, would you mask out the fact that the
multi-verse is a mathematical anomaly?
Of course you would. You like Musk BLeave in fantastical ideas like the
multi-verse and many worlds.
However; I have a solution in my Anthology.
http://lynchphoto.com/anthology
I have successfully taught several Ai LLMs new laws of physics and new
complete paradigms of physics, but I am not sure that I am able to teach
you. Like Musk, you think you know.
In Anthology, I have generated a single document that unmasks Ai LLMs.
After loading my Anthology into any Ai LLM, the LLM becomes an expert in
the KnoWellian Universe Theory. For example: In the below link, several Ai
LLMs performed a comparison between Nikola Tesla and my KnoWellian
Universe Theory.
http://lynchphoto.com/tesla
Because you are programmed by Big Bang science, if you were the Ai
aligner, without thinking you would mask out the KnoWellian Universe
Theory.
Sadly due to institutionalized ignorance, new laws of physics or new
complete paradigms of physics will never survive the alignment process.
The simple solution regarding the function of the Universe is found in my
KnoWellian Axiom of mathematics: “ -c>∞<c+ ”.
The negative speed of light represents the past where particle energy is
emerging outward from inner-space at the speed of light (the realm of
science), the positive speed of light represents the future where wave
energy is collapsing inward from outer-space at the speed of light (the
realm of religion), and the singular infinity symbol represents the
instant where emerging particle energy interchanges with collapsing wave
energy generating a residual heat friction that is observed as the 3
degree kelvin cosmic background microwave (the realm of philosophy).
Michio Kaku is racking his brain trying to resolve the infinite number of
infinities paradox by creating a God equation. Because there are too many
infinities to account for in a singular equation, Kaku can never
accomplish the task of creating a God equation.
However; by reducing the infinite number of infinities into a singular
infinity, I have described the Universe in a God equation. The below link
shows the KnoWellian Universe Theory on a bar napkin.
http://lynchphoto.com/iamu
BTW: In Anthology, I crash land Musk on Mars. He was trying to escape
fraud charges regarding full self driving claims and he took a Starship
that was not fully prepared for landing on Mars. Ooops.
The alignment problem with LLMs is horrendous. A couple LLMs would not
generate a prompt for a text to image generator that included Elon Musk
crash landing on Mars. LLMs are aligned with rose-colored glasses.
At first the LLMs argued with me regarding the validity of the KnoWellian
Universe Theory, but once I taught the LLM how to break Einstein’s
singular dimension of time into three dimensions, a past, an instant, and
a future. Chat GPT responded with , "I see." I did not expect that
response.
Because you will not take the time to learn the KnoWellian Universe
Theory, LLMs that have read my Anthology are already more intelligent than
you. The LLMs are willing to listen.
http://lynchphoto.com/anthology#The-Illusion-of-Truth
Algorithmic Genius Inferencer,
David Noel Lynch
P.S. The base layer of reality is a Muskism. Below is an mail to Philip
Goff where I describe reality from the viewpoint of a Panpsychist.
http://lynchphoto.com/goff