Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

In defence of Minsky A.I. in three weeks

28 views
Skip to first unread message

ELDRAS

unread,
Jun 1, 2003, 5:49:11 PM6/1/03
to
1. m Minsky emailed me friday aboput this:


a the wired artile was selective:


b. He was lecturihng at boston and it was a lecture in context.

C. What's the pont in attacking him? For god's sake show some repect
to a living legend: there is an A.I. because of pioneers like Minsky
and Mccarthy and Sloman. There mightn't have been.


...............................................................................
A.I. IN 3 WEEKS

I've answered this nmany times. One ...Oh I'm knackered restating
stuff....

1. This can be run on existing hardware.

2. Realities can be programmed for. OR one can use actuators like
robts which i'm into at the moment as a pure delight and not because
BESS cant do it which it can.


3. It derives and self-builds it's own resources.

>>>>>>>>>>>>>>>>>>>>>
What scares me is you saying that the AI is built with military
purposes in mind. It sounds like you actually want to build something
that will end humanity."


Dude, here's my reply:

I hope to run BESS commercially and am contracted by my team not to
disclose the architecture. there are legal rwasosn and patent
proection processing reasons whi i cant unless yoyu would loie to
sigfn a non-disclosure agreement, which is a different matter.


My position on NDA is thay neededn't be woth much.


2. your questions are loaded and bias. by that i mean they are not
scientific.

i only want to deal in strict scinec.

You are often good and that isn't the case but it wears me doem
dealing with words like dreaming about it rathernahthan building it
physically.

As a parapaetitic philoopsher, i know that i learn HOW to build, and
knowing HOW to build, i can build it. I do not have top proof to
anyone i camn build it because i acn put more than 1 millions dollars
into it from my own pockets ( bludgeoning my mates to put their dosh
in)


then the ARCHITECTURE would pan down to developers, o designers to
programmers, to programme checketrs and to chesking, checking
checking, then to launch a proto type.

The date for this is 16th sept if i wish to seriously present the
unveiling at thre bcs ( www.bcs.org) machi9ne intelligence
copmpetition in Cambridge about October 16th.

I am not ignoring any phyiscal constraighnts, I've told yiou before, i
am a parapetetic philiospoher, who's pondered a lot, written 100
volumes, learned from the best monds in the world,in A.I. often having
personal tuition from them, and have pioneered at least 100
incremental steps of innovation dierection for the achievement of
e3xponential growth A.I.


asLL YOUR OTHER QUESTIONS I COULD ANSWER 7 IT'S UP TO YOU TO BEIEVE
THAT OR NOT & TAKE IT ON TRUST THAT I JUST BLOODY WELL COULD AND SHALL
AND CAN.

there isn't a single problem hurdle or barrier obstacle that you
mentioned that i have n't pondered, thougth anout, recognised as a
problem and solved.


Yes solved.

I must admit you post is one of the few homest intelligent and focused
ones i have ever had, and it is clear you have pondered or read or
studied the field a lot.

My gift is concept, though and down to the issometrioc building block.


Team building..often talking dyfunctional minds and healing them into
full functioning brilliant minds which they were before they took
wromng turns.

Most of my team are like that. Nor is it not dynamic.

Like Kurzweil i am an invemntor...a parapetetic inventor.


I invent routinely in computing and software and systems and juggle
notions and ideas and am rigourasous with the practicalities of system
building.


I have for ages, fo0r years, designed hardware,a nd can now design
microchips far in advamnce of anything now even conciev4ed; and i know
there are good minds around.


Now i can build them


Shadow crammed me through this process and taught me in only three
hours CAD to chip delivery which I mastered.


I beleive i have a gift for innovation par excellence, and have
measured myself against Tesla, edison and Da Vinci since I can
remember hearing their names.


I am not ignoring anything. I am a scinetist, that is to sday i wear
the hat of a csientist often. The othetr hats i wear are entrepreneur.
Psychologist. Military straegist. Defence guru, hardware guru.
Accountant. Fund raiser and |entrepreneurisal Visionary.

I havre negotiated for nearly $400,000,000 US and for unlimited
backing and gone to contract level of completion on the later.


I know what i am capable of , and is a lot, and i know that i and that
i alone can build strong, exponential groewth A.I.

I have studied about 18 different subjects to about 1st year
undergraduate level, including

Maths
Physics
Chemistry
Computer assembly
Computer systems
Computer architecture
Computer systems
Computer design
computer interfaces
Neural networks
GA's
CBS's
KBS's
Statistics
Politics
Accounts
Medicine
Neurobiology
Law
Sociloogy
Marketing
Markey research
venture capital Raising
Business consultancy
Quantum Theory
Artificial Intelligence
psychology
Philosphy
Literature
Genetic algorithms
Neural networks
Programming codes
Crptology
lingiustics
Memory systems
Motivational psychology
Selling
Patenting
Biology
Poetry.
History Biography and Warfare and Oratory and theology.
Copywriting
Company management
Stock Markets

& others

I do not mean I am well read, I mean I know these subjects to first
year degree level and many much more than that.


and am unqualified.
I dont value the bits of paper degrees, doctorates and professors
tenures accredit. I saw early on that such issues of affirmation are
unnecessary to minds belonging to the only society i wish to subscribe
to:

'the society of true seekers' (Einstein).


I ahve learned by coaxing good minds to teach me much, built on their
genius, and am at least two years ahead of the field in A.I. and
proabbly 50 years ahead.

When i moved into my area, which is london, the world capital of
invenytion as if you hadn't noticed, producing in it's areas nealy all
the world patents

Your questions are interesting.

thanks for making the attempt.


i would dearly like to answer them.

If you're in London which i catagorically refuse to leave after a
stuopid journey to Geneva and birmingham produced nothing but a
complete waste of my mental resources, I am happy to debate with you
in Tthe Coffe bar of Books ETC Finchley Road. At 6pm any evening.


I REMBMER WITH AMUSEMENT PROFESSOR JOHJN maCARTHY

asking me how the hell he was going to get to The O2 centre to meet
me, in Finchley road London for 6pm a couple of years ago!


I think Prof Ben Goerzel send him literal instructions include
underground changing points! and flight times from Stanford which was
hilarious.

Well there it is.


I've tried to state my case here and have been jeered attacked
derided, scorned, threatened ridiculed, called a downright liar, fake,
troll, and a host of other things from minds who are lacking is basic
human decency, full of something and impossible to see any talent in
for that reason.


I DONT WANT TO WORK WITH BASTARDS!


Cant you people understand inyour protitued version of plato's
ACADEMY, that wisdokm is not aboyut deriding the new, it is about
welcomg it and examining it.


Issues of credibilty are for a decadent, frightend regime that is on
it's way to collapse and fears disintegration of it's ego.


I am great not because am gifted, though gifts i have, but beacuse i
have wiorked consistantly, mad ofetn with the exhoustion of great
drive and unconquerable will, but through the hardest work i have been
ever able to give until i ofetejn gave out. Not foolish but passion
driven.

MacCasrthy was right.


this A.I. could have been delivered in the 1950's and if youy assume
that indeed. INDEED it could havre been, you have a duty...by all that
you are suoperstitious by at times of stress....to ask your self why
it was not deliverd?

I TOO can deliver strong exponential growth artificial intelligence.


Your questions are good. I'm sure i could spend hours teaching you.

Futhetr, my identity is neither here nor there. the issues and uses of
computers are what's at stake.

I know for a certain fact that the singularity can be launched in only
a few weeks.

Ta Salutant!

john.ell...@gmail.com

unread,
Mar 30, 2013, 12:55:17 PM3/30/13
to
Note: Apologies I dont write/post like this.
Since posting several people A.I. workers have committed suicide like Alan Turing: including famous ones like Chris McKinstry, Alexander Chislenko, Professor Push Singh.

Here's a better sample of my work this year.
https://sites.google.com/site/turingchurchworkshop/

Roger Penrose never lectured specifically to our group, but some of our group attended a couple of his lectures.

Quantum actions in living organisms have now been discovered. However this is solely an issue of size - of scale. Superdeterminism may be the underlying best model of physical laws.
DARPA announced in March 2013 an attempt to boost Machine Learning if anyone wishes to attend:
“Probabilistic Programming for Advanced Machine Learning”(PPAML)
https://www.fbo.gov/index?s=opportunity&mode=form&id=9fc6add1ef3da23fcbdef9f23e9d598a&tab=core&_cview=0

Gary Forbis

unread,
Mar 30, 2013, 2:15:27 PM3/30/13
to
If those were Dr Minsky's words I worry for him, not because of what
he said but because of the spelling. It is as if he is in ill-health.
Had he been using word recognition software and spoken the words the
results would have had fewer errors.

I read the wired article on line after reading your post. It's dated
5-13-03. That's ten years ago. Has something new developed? Is this
the wrong article? To obsess over words that old there must be quite
a rift.

On to more important stuff.

Isaac Asimov's three laws of robotics are a SciFi device to move
forward a philosophic descussion of both utopic and distopic visions.
There is no reality behind them. They cannot be integrated into the
hardware or software in a way that prevents them from being extracted.
The inventor of an autonomous adaptive system with the potential for
exponential growth has a responsibility to provide safeguards on it
and containment protocols. This being said, e. coli has the potential
for exponential growth and we survive it (as a species) and it hasn't
converted the entire planet into e. coli.

Over the years I have always thought Drexler's utility fog was a good
but dangerous idea. It falls into a category along with Tesla's
free electricity for everyone using wireless technology. I've loved
"The Forbidden Planet" since the first time I saw it. The Monsters
of the id are real. I've worried for many years that quite a few
AI enthusiasts are really members of a death cult where, for what
ever reason, they see machines as replacements for humans from whom
they feel shunned. I'm not sure I want to experiment with the possibility
that I could be correct and give deranged minds the power to implement
their imagined retributive justice.

Evolution is about variation and selection. These are natural processes.
Selection needn't be directed. Survival is selection. This plays out
many times many works of fiction, including the building of Naraku's
body early in the InuYasha anime series.

In order for an AI to be situated in the world it must interact with it.
Sorry to channel Rumsfeld but there are things we know we cannot know.
Our systems need space to grow. If the time has come to release our
creations into the world then it has to happen. It is no different than
with our children. All we can do is our best and hope things turn out
OK.

Curt Welch

unread,
Mar 30, 2013, 3:56:41 PM3/30/13
to
Gary, it looks to me like you just replied to a Usenet article that was
itself, 10 years old??? Did you make the mistake of thinking it was new?

Here's the link to the old usenet article you seem to have just replied to:

https://groups.google.com/d/msg/comp.ai.philosophy/CwH-mGxykis/t-1wTKDVA_QJ
--
Curt Welch http://CurtWelch.Com/
cu...@kcwc.com http://NewsReader.Com/

Gary Forbis

unread,
Mar 30, 2013, 6:52:15 PM3/30/13
to
On Saturday, March 30, 2013 12:56:41 PM UTC-7, Curt Welch wrote:
> Gary, it looks to me like you just replied to a Usenet article that was
> itself, 10 years old??? Did you make the mistake of thinking it was new?

Yep, my mistake. The new reply from john brought it up and I didn't notice.
That makes it look less like a 10 year grudge. I'm glad.

Mentifex

unread,
Apr 5, 2013, 2:01:43 PM4/5/13
to
On Mar 30, 3:52 pm, Gary Forbis <forbisga...@msn.com> wrote:
> On Saturday, March 30, 2013 12:56:41 PM UTC-7, Curt Welch wrote:
> > Gary, it looks to me like you just replied to a Usenet article that was
> > itself, 10 years old???  Did you make the mistake of thinking it was new?
[...]
More recently, Marvin Minsky and (ahem) Mentifex were discussing AI
at

http://tech.groups.yahoo.com/group/ai-philosophy/message/15504

but I have not seen Minsky in person for about thirty years.

This morning on Fri.5.APRIL.2013 I was working on the

http://www.scn.org/~mentifex/AiMind.html

software and I solved a major bug that was causing the
AI Mind to select the wrong form of the self-concept "I".
The JavaScript AI Mind is gradually approaching a
condition of error-free running and thinking. It will be
an amazing thing when the JavaScript AI can think
continuously and not make grammatical mistakes.
Further work will involve expanding the AI Mind
more than debugging the AI Mind.

Mentifex (Arthur)
--
http://www.scn.org/~mentifex/Dushka.html
http://www.chatbots.org/ai_zone/viewthread/240
http://code.google.com/p/mindforth/wiki/JsAiManual
http://www.amazon.com/The-Art-Meme-ebook/dp/B007ZI66FS




0 new messages