Foundations of Mind V "The new AI scare" proceedings now online

25 views
Skip to first unread message

University of Ireland

unread,
Jan 14, 2018, 4:28:39 AM1/14/18
to Online_Sa...@googlegroups.com
http://cosmosandhistory.org/index.php/journal/issue/current

24 new papers including Jack Sarfatti, Chris Langan and many members
of this list

Please enjoy!

Paul Werbos

unread,
Jan 14, 2018, 2:35:57 PM1/14/18
to Online_Sa...@googlegroups.com
So Jack Sarfatti will explain to us what AI is, how it works, and what it may do?



--
----------------------------
Fifth International Conference
Science and Scientist - 2017
August 18—19, 2017
Nepal Pragya Pratisthan, Kathmandu, Nepal
http://scsiscs.org/conference/scienceandscientist/2017

Send a Donation to Support Our Services: http://scienceandscientist.org/donate
(All Indian residents are eligible for tax benefits for their contributions under section 80G of the Income Tax Act)

Report Archives: http://bviscs.org/reports

Why Biology is Beyond Physical Sciences?: http://dx.doi.org/10.5923/j.als.20160601.03

Life and consciousness – The Vedāntic view: http://dx.doi.org/10.1080/19420889.2015.1085138

Harmonizer: http://scienceandscientist.org/harmonizer

Darwin Under Siege: http://scienceandscientist.org/Darwin

Princeton Bhakti Vedanta Institute: http://bviscs.org

Sri Chaitanya Saraswat Institute: http://scsiscs.org

Sadhu-Sanga Blog: http://mahaprabhu.net/satsanga

Contact Us: http://scsiscs.org/contact
---
You received this message because you are subscribed to the Google Groups "Sadhu-Sanga Under the holy association of Spd. B.M. Puri Maharaja, Ph.D." group.
To unsubscribe from this group and stop receiving emails from it, send an email to Online_Sadhu_Sanga+unsub...@googlegroups.com.
To post to this group, send email to Online_Sadhu_Sanga@googlegroups.com.
Visit this group at https://groups.google.com/group/Online_Sadhu_Sanga.
To view this discussion on the web visit https://groups.google.com/d/msgid/Online_Sadhu_Sanga/CAJbqt2EWnSb7Hf6C2iTF6CiULVPrGcmYVrNebj-fiOeJSRbUpQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

University of Ireland

unread,
Jan 16, 2018, 4:21:59 AM1/16/18
to Online_Sa...@googlegroups.com
On 1/14/18, Paul Werbos <paul....@gmail.com> wrote:
> So Jack Sarfatti will explain to us what AI is, how it works, and what it
> may do?

Jack is one of 21 contributors and has a unique take on how Bohmian
(post) QM relates to subjectivity

As it happens, I am probably the only comp sci Ph.D. in the 21. It is
my view that this new scare was hustled up by Tegmark/Musk etc to
secure new sources of funding as AI is still very primitive

Best

S
>> email to Online_Sadhu_Sa...@googlegroups.com.
>> To post to this group, send email to Online_Sa...@googlegroups.com.
>> Visit this group at https://groups.google.com/group/Online_Sadhu_Sanga.
>> To view this discussion on the web visit https://groups.google.com/d/
>> msgid/Online_Sadhu_Sanga/CAJbqt2EWnSb7Hf6C2iTF6CiULVPrG
>> cmYVrNebj-fiOeJSRbUpQ%40mail.gmail.com.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
> email to Online_Sadhu_Sa...@googlegroups.com.
> To post to this group, send email to Online_Sa...@googlegroups.com.
> Visit this group at https://groups.google.com/group/Online_Sadhu_Sanga.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/Online_Sadhu_Sanga/CACLqmgct%2B3zBonw_xCKHLw%3Ddng13ATUXCaCCMid%3Dk46euoFgvQ%40mail.gmail.com.

Whit Blauvelt

unread,
Jan 16, 2018, 1:05:23 PM1/16/18
to Online_Sa...@googlegroups.com
On Mon, Jan 15, 2018 at 07:49:46PM -0800, University of Ireland wrote:
>
> It is my view that this new scare was hustled up by Tegmark/Musk etc to
> secure new sources of funding as AI is still very primitive

Amusing, the notion that Musk would have to scare people to hustle up
funding, when it seems that money just flows to him. Here I'd though I'd
bought stock in his enterprises because he builds stuff that works. In any
case, he's far more aligned to promise than fear in the popular imagination.

Best,
Whit

NYIKOS, PETER

unread,
Jan 16, 2018, 2:54:00 PM1/16/18
to Online_Sa...@googlegroups.com
I think you are both right, each in his own way. Grant applicants are viewed favorably by people who are close enough to them so that their own proposals are looked upon favorably. So AI researchers can expect to overstate the case when reviewing AI papers on behalf of the granting agency, while still sounding objective.

Popularizers are especially important to granting agencies, which in turn are dependent on stockholders or the legislative branches of the government. Each agency needs to be able to sell the ultimate source on the importance of the things being funded. If the public as a whole got wind of the primitive current state of AI, the agencies might shift their funding in more promising directions.

Scare tactics are very effective in selling the general public -- just look at all the hype about "tipping points" in global warming. Responsible climatologists know that we have no idea what this tipping point is for a positive feedback loop to produce runaway warming, nor how far the runaway can go before new factors enter into the equation, as happened after the Paleocene Maximum and also in the Miocene as the Antarctic ice sheet began growing again after it melted temporarily.

Such facts do nothing to discredit those, like Al Gore, who made definite predictions which have not materialized years after they were supposed to have taken place. The public is still made to think that we are in a perpetual state of emergency where global warming is ooncerned.


Peter Nyikos
________________________________________
From: online_sa...@googlegroups.com [online_sa...@googlegroups.com] on behalf of Whit Blauvelt [wh...@csmind.com]
Sent: Tuesday, January 16, 2018 10:29 AM
To: Online_Sa...@googlegroups.com
Subject: Re: [Sadhu Sanga] Foundations of Mind V "The new AI scare" proceedings now online

Best,
Whit

--


----------------------------
Fifth International Conference
Science and Scientist - 2017
August 18—19, 2017
Nepal Pragya Pratisthan, Kathmandu, Nepal
http://scsiscs.org/conference/scienceandscientist/2017

Send a Donation to Support Our Services: http://scienceandscientist.org/donate
(All Indian residents are eligible for tax benefits for their contributions under section 80G of the Income Tax Act)

Report Archives: http://bviscs.org/reports

Why Biology is Beyond Physical Sciences?: http://dx.doi.org/10.5923/j.als.20160601.03

Life and consciousness – The Vedāntic view: http://dx.doi.org/10.1080/19420889.2015.1085138

Harmonizer: http://scienceandscientist.org/harmonizer

Princeton Bhakti Vedanta Institute: http://bviscs.org

Sri Chaitanya Saraswat Institute: http://scsiscs.org

Sadhu-Sanga Blog: http://mahaprabhu.net/satsanga

Contact Us: http://scsiscs.org/contact
---
You received this message because you are subscribed to the Google Groups "Sadhu-Sanga Under the holy association of Spd. B.M. Puri Maharaja, Ph.D." group.
To unsubscribe from this group and stop receiving emails from it, send an email to Online_Sadhu_Sa...@googlegroups.com.
To post to this group, send email to Online_Sa...@googlegroups.com.
Visit this group at https://groups.google.com/group/Online_Sadhu_Sanga.

To view this discussion on the web visit https://groups.google.com/d/msgid/Online_Sadhu_Sanga/20180116152908.GB24577%40black.transpect.com.

University of Ireland

unread,
Jan 17, 2018, 4:02:07 AM1/17/18
to Online_Sa...@googlegroups.com
Yes, somebody likes Elon a lot. Dave Matthews, Elon, Henry Markram
etc; a new operation Paperclip, this tome from SA

as I understand it, his last rocket failed - as many before

What is unsustainable for the US is non-coders wading into AI - much
like the disastrous Henry in the EU brain project

Just because someone is arrogant doesn't mean he's competent
> https://groups.google.com/d/msgid/Online_Sadhu_Sanga/20180116152908.GB24577%40black.transpect.com.

Paul Werbos

unread,
Jan 17, 2018, 7:18:25 AM1/17/18
to Online_Sa...@googlegroups.com
On Tue, Jan 16, 2018 at 11:54 PM, University of Ireland <universit...@gmail.com> wrote:
Yes, somebody likes Elon a lot. Dave Matthews, Elon, Henry Markram
etc; a new operation Paperclip, this tome from SA

as I understand it, his  last rocket failed - as many before

What is unsustainable for the US is non-coders wading into AI - much
like the disastrous Henry in the EU brain project


I too have mixed feelings about Elon Musk -- but then again there are few people I view in purely black or white terms. 

But Jack Sarfatti's evaluation of an issue which Musk has glommed onto is hardly worthy of our time.

As it happens, the threat of Terminator AI is far more real than most people know, even the computer scientists trained before the "new AI deep learning revolution." The "friendly AI" solutions which some folks have funded are basically laughable, closer than you would believe to using crayons to draw happy faces on killer drones unaffected by such crayon markings.

In December 2016, I was asked to give the keynote plenary to the world's top technical meeting on the new AI. The guy who invited me was appalled to hear that exactly his algorithm is now being used for killer drones to decide whom to kill, autonomously. Just a few questions" do we want to make such algorithms more intelligent and focused, or more stupid and random? Who sets the values and how?




 
> email to Online_Sadhu_Sanga+unsub...@googlegroups.com.
> To post to this group, send email to Online_Sadhu_Sanga@googlegroups.com.
--
----------------------------
Fifth International Conference
Science and Scientist - 2017
August 18—19, 2017
Nepal Pragya Pratisthan, Kathmandu, Nepal
http://scsiscs.org/conference/scienceandscientist/2017

Send a Donation to Support Our Services: http://scienceandscientist.org/donate
(All Indian residents are eligible for tax benefits for their contributions under section 80G of the Income Tax Act)

Report Archives: http://bviscs.org/reports

Why Biology is Beyond Physical Sciences?: http://dx.doi.org/10.5923/j.als.20160601.03

Life and consciousness – The Vedāntic view: http://dx.doi.org/10.1080/19420889.2015.1085138

Harmonizer: http://scienceandscientist.org/harmonizer

Darwin Under Siege: http://scienceandscientist.org/Darwin

Princeton Bhakti Vedanta Institute: http://bviscs.org

Sri Chaitanya Saraswat Institute: http://scsiscs.org

Sadhu-Sanga Blog: http://mahaprabhu.net/satsanga

Contact Us: http://scsiscs.org/contact
---
You received this message because you are subscribed to the Google Groups "Sadhu-Sanga Under the holy association of Spd. B.M. Puri Maharaja, Ph.D." group.
To unsubscribe from this group and stop receiving emails from it, send an email to Online_Sadhu_Sanga+unsub...@googlegroups.com.
To post to this group, send email to Online_Sadhu_Sanga@googlegroups.com.

Whit Blauvelt

unread,
Jan 17, 2018, 11:44:29 AM1/17/18
to Online_Sa...@googlegroups.com
On Tue, Jan 16, 2018 at 07:06:34PM +0000, NYIKOS, PETER wrote:

> Scare tactics are very effective in selling the general public -- just
> look at all the hype about "tipping points" in global warming.
> Responsible climatologists know that we have no idea what this tipping
> point is for a positive feedback loop to produce runaway warming, nor how
> far the runaway can go before new factors enter into the equation, as
> happened after the Paleocene Maximum and also in the Miocene as the
> Antarctic ice sheet began growing again after it melted temporarily.

Peter,

On the contrary, scare tactics have not worked, as is evidenced by the
election of a national government in the US which discounts all science.
People so dislike being scared they're willing to discard their former faith
in science in order to shrug off the fear.

What's needed to avoid the threats from an overly-carbonized atmosphere is
more emphasis on the superior prospects of a green future. I was born in the
50s. I rememeber when American culture was informed more by positive,
science-fictional dreams than by dystopias. That's where Musk gets it. His
electric cars, his solar rooftops, his superior batteries, his reusable
rockets, his hyperloop rapid transit concept, even his colonization of Mars,
while all recognizing the reality of threats to civilization, are foremost
hopes, motivating by their beauty, not so much by fear of the alternatives.

Now General Motors plans to transition to an all-electric vehicle fleet.
That's not because GM is scared of carbon; it's because they realize that
Musk's positive dream of electric vehicles will carry the day by its beauty.
At the same time, solar and wind energy production are now cheaper than
fossil fuels in much of the world. Without government favoritism and
subsidy, fossil fuel production is a fading industry.

What scare tactics do, perversely, is turn people conservative. When
psychologists give people, whose politics are unknown, assessments of how
fearful those people are, psychologists can accurately predict their
politics: those who have the most fear are the most conservative, seeking
safety in familiar forms. And increasing fear in people, in itself, tilts
them towards conservative stances, even when the fear and the stances are
about separate issues. That's why certain politicians base their campaigns
on fear of the Other -- other races, nationalities, religions.... The most
important part of the message, the most politically productive, is the fear
itself.

So yeah, I agree, scientists promoting fear is not the effective way to get
people to go postively into the future, or even get societies to avoid the
things that are the true dangers to us.

Best,
Whit

Menas Kafatos

unread,
Jan 17, 2018, 11:44:29 AM1/17/18
to Online_Sa...@googlegroups.com
Yeah, Paul is right. As is Sean. Different agendas of many “experts” on AI.


Sent from my iPhone
To unsubscribe from this group and stop receiving emails from it, send an email to Online_Sadhu_Sa...@googlegroups.com.
To post to this group, send email to Online_Sa...@googlegroups.com.

NYIKOS, PETER

unread,
Jan 18, 2018, 11:45:09 AM1/18/18
to Online_Sa...@googlegroups.com
There are lots of computer advances for peace-loving people to be appalled at, but which ones really deserve to be called AI?

Killer drones are, at besr, very low-tech AI. And perhaps the cyberneticists in the 1950's were also "appalled" at antiaircaft rockets that guided themselves either by the heat of the target or its radar signature. And those of the late 60's may similarly have been appalled at "smart bombs" like the one that destroyed a North Vietnamese bridge that had survived many earlier "dumb" bombing attempts.


From: online_sa...@googlegroups.com [online_sa...@googlegroups.com] on behalf of Paul Werbos [paul....@gmail.com]
Sent: Wednesday, January 17, 2018 5:30 AM
To: online_sa...@googlegroups.com
Subject: [MaybeSpam]Re: [Sadhu Sanga] Foundations of Mind V "The new AI scare" proceedings now online

Whit Blauvelt

unread,
Jan 18, 2018, 2:11:46 PM1/18/18
to Online_Sa...@googlegroups.com
On Thu, Jan 18, 2018 at 04:08:33PM +0000, NYIKOS, PETER wrote:
> There are lots of computer advances for peace-loving people to be appalled at,
> but which ones really deserve to be called AI?
>
> Killer drones are, at besr, very low-tech AI. And perhaps the cyberneticists in
> the 1950's were also "appalled" at antiaircaft rockets that guided themselves
> either by the heat of the target or its radar signature. And those of the late
> 60's may similarly have been appalled at "smart bombs" like the one that
> destroyed a North Vietnamese bridge that had survived many earlier "dumb"
> bombing attempts.

Hi Peter,

Many of my friends are serious pacifists. I don't know of anyone being
appalled at more accurate bombs, since the alternative is more collateral
damage. More accurate anti-missile programs, by contrast, have faced some
resistance in the peacenik community, but largely because they both
encourage the devising of even more devastating offensive weapons to counter
them, and because they often don't really perform up to spec, creating a
false sense of security that encourages use of offensive weapons.

Why do you claim killer drones are "low-tech AI"? From what we're spending
on them, and considering the considerable military-funded research in AI,
what's your knowledge that it's not incorporated in our drones? A relative
of mine is an aerospace engineer with a major defense contractor; the talent
we've invested there is not second-rate.

Self-driving cars do complex stuff. Let's grant that current self-driving
drones are farther ahead on the development curve. The cars want to keep
everyone alive, while the drones want to selectively cull the population.
Whether or not those drones "deserve" to be called "AI" as their develoment
advances -- whatever we want to call them -- they're going to be dangerous.

Assuming there's no breakthrough in bringing consciousness to AI, they're
also going to be the ultimate in "cold blooded killers." Whether that's a
difference that makes a difference -- the consciousness or lack in the drone
that's about to kill you -- is still an open question.

Best,
Whit

Paul Werbos

unread,
Jan 18, 2018, 3:56:03 PM1/18/18
to Online_Sa...@googlegroups.com
On Thu, Jan 18, 2018 at 1:59 PM, Whit Blauvelt <wh...@csmind.com> wrote to Peter:

Why do you claim killer drones are "low-tech AI"? From what we're spending
on them, and considering the considerable military-funded research in AI,
what's your knowledge that it's not incorporated in our drones?

Indeed. Whit is right. I am amazed at how much confidence so many people have in what they think they know, in areas they know nothing about,
If I had had time, I would have headed  my recent book chapter with a quote from Mark Twain, "It's not what you don't know that kills you, it's 
(what you know that isn't so)."

In the case of the drone, I know exactly what was in it. The algorithm was the first one featured in the advanced symposium for the next major wave in deep learning and AI, at the top meeting in the world on that subject. I thought I said that already, but clearly I need to work on my very limited communication skills. Sorry.

Best of luck,

    Paul

P.S. Guess why I don't try to elaborate further and give more detailed instructions for how to wipe out the human species? 

But there are people much better placed than I am politically who accuse me of being a "racist" -- human racist. "Who needs humans? Why are you prejudiced against silicon-based life forms?"

University of Ireland

unread,
Jan 18, 2018, 6:44:11 PM1/18/18
to Online_Sa...@googlegroups.com
>
> Indeed. Whit is right. I am amazed at how much confidence so many people
> have in what they think they know, in areas they know nothing about,
> If I had had time, I would have headed my recent book chapter with a quote
> from Mark Twain, "It's not what you don't know that kills you, it's
> (what you know that isn't so)."
>
> In the case of the drone, I know exactly what was in it. The algorithm was
> the first one featured in the advanced symposium for the next major wave in
> deep learning and AI, at the top meeting in the world on that subject

I repeat - Jack's paper is about the Bohmian physics in which he is
genuinely expert, to the point he was invited to speak at Bohm's
centenary in London

Secondly, beautiful book about the military abuse of drones;

https://www.amazon.com/Kill-Chain-Rise-High-Tech-Assassins/dp/1250081637

Thirdly, forgive my deep skepticism about "the top meeting in the
world on that subject"

This week, such a meeting on the brain at Cal is utterly ignorant
about Hodgkin-Huxley, Freeman, and indeed can be justly called
"Neurostatics;


https://simons.berkeley.edu/workshops/brain2018-boot-camp

. I
> thought I said that already, but clearly I need to work on my very limited
> communication skills. Sorry.
>
> Best of luck,
>
> Paul
>
> P.S. Guess why I don't try to elaborate further and give more detailed
> instructions for how to wipe out the human species?
>
> But there are people much better placed than I am politically who accuse me
> of being a "racist" -- human racist. "Who needs humans? Why are you
> prejudiced against silicon-based life forms?"
>
> https://groups.google.com/d/msgid/Online_Sadhu_Sanga/CACLqmgdDDj%3DNSNkHJuiOMJHxu-FdTDQZXZhU%2BYV28yoGHoVbOg%40mail.gmail.com.

Paul Werbos

unread,
Jan 18, 2018, 8:39:08 PM1/18/18
to Online_Sa...@googlegroups.com
On Thu, Jan 18, 2018 at 5:56 PM, University of Ireland <universit...@gmail.com> wrote:
>
> Indeed. Whit is right. I am amazed at how much confidence so many people
> have in what they think they know, in areas they know nothing about,
> If I had had time, I would have headed  my recent book chapter with a quote
> from Mark Twain, "It's not what you don't know that kills you, it's
> (what you know that isn't so)."
>
> In the case of the drone, I know exactly what was in it. The algorithm was
> the first one featured in the advanced symposium for the next major wave in
> deep learning and AI, at the top meeting in the world on that subject

I repeat - Jack's paper is about the Bohmian physics in which he is
genuinely expert, to the point he was invited to speak at Bohm's
centenary in London

Gee. I was not even thinking about Jack in what I posted, and I was worried about how I was not tactful enough to the guy Whit was replying to. I am glad to hear people are not pointing back to him (the other guy).

It is interesting how you immediately thought about Jack, when I referred to folks making strong statements about 
the new AI without knowing about THAT.
 
Secondly, beautiful book about the military abuse of drones;

https://www.amazon.com/Kill-Chain-Rise-High-Tech-Assassins/dp/1250081637

Thirdly, forgive my deep skepticism about "the top meeting in the
world on that subject"


Of course I respect skepticism. For tangible evidence, see the top of my "Mind" web page and the links it congtains:

 ==============================================

January 2017: Sergey Brin and his top people in “Deep Mind” have spoken recently about new breakthroughs and the coming big new waves in deep learning and new “AI.” For a brief overview, a link to his recent talk and a link to my own lead talk to the big symposium on that topic last month, click here. (For slightly expanded version of that talk, click here.)


=======================================================

 
This week, such a meeting on the brain at Cal is utterly ignorant
about Hodgkin-Huxley, Freeman, and indeed can be justly called
"Neurostatics;


It sounds as if I would agree with you in general terms. AI and brain studies have strong logical connections, but I would not characterize a narrow topical meeting in neuroscience with the technology of the new AI.
 

https://simons.berkeley.edu/workshops/brain2018-boot-camp

AI had a great cultural revolution starting in 2008 or 2009 or so, driven by new empirical result. The empirical results this past year or two in neuroscience strongly support the claims by Freeman, Pribram and others of us that new paradigms really are justified now in that field too,  but  I would have said that by 2000 about the evidence accumulated by then in AI. Human cultures are often a lot slower and more rigid than the foundations of the scientific method call for. But... well.. just part of the normal process of evolution.

Best of luck,

    Paul 

University of Ireland

unread,
Jan 19, 2018, 6:50:12 AM1/19/18
to Online_Sa...@googlegroups.com
thanks for the thoughtful reply, Paul

I add the original message;

"http://cosmosandhistory.org/index.php/journal/issue/current

24 new papers including Jack Sarfatti, Chris Langan and many members
of this list

Please enjoy!"
> ==============================================January 2017: Sergey Brin
> and
> his top people in “Deep Mind” have spoken recently about new breakthroughs
> and the coming big new waves in deep learning and new “AI.” For a brief
> overview, a link to his recent talk and a link to my own lead talk to the
> big symposium on that topic last month, click here
> <http://drpauljohn.blogspot.com/2017/01/deep-learning-and-new-ai-you-aint-seen.html>
> . (For slightly expanded version of that talk, click here
> <http://www.werbos.com/Neural/AI_keynote_review.pdf>.)
>
> =======================================================
>
>
>
>> This week, such a meeting on the brain at Cal is utterly ignorant
>> about Hodgkin-Huxley, Freeman, and indeed can be justly called
>> "Neurostatics;
>>
>>
> It sounds as if I would agree with you in general terms. AI and brain
> studies have strong logical connections, but I would not characterize a
> narrow topical meeting in neuroscience with the technology of the new AI.
>
>
>>
>> https://simons.berkeley.edu/workshops/brain2018-boot-camp
>
>
> AI had a great cultural revolution starting in 2008 or 2009 or so, driven
> by new empirical result. The empirical results this past year or two in
> neuroscience strongly support the claims by Freeman, Pribram and others of
> us that new paradigms really are justified now in that field too, but I
> would have said that by 2000 about the evidence accumulated by then in AI.
> Human cultures are often a lot slower and more rigid than the foundations
> of the scientific method call for. But... well.. just part of the normal
> process of evolution.
>
> Best of luck,
>
> Paul
>
> https://groups.google.com/d/msgid/Online_Sadhu_Sanga/CACLqmgfkb6QsJGkyHUKTDsoOBso%3DcBNMfG%2Bh_64Ci0UDu4M0EA%40mail.gmail.com.
Reply all
Reply to author
Forward
0 new messages