Inkwell: Authors and Artists

    

I find that I fall back on their attitudes, almost unthinkingly,
when I'm confronted by some intractable, wicked, long-term problem
where every moral trolley-switch kills somebody.   "You think that
crap impresses me?!  I've seen so much worse!  It'll pass!" 
Commonly followed, in their case, by a pack of cigarettes and half a
bottle of fruit brandy, but if you're watching yourself, you don't
have to perform that part.

Also, Russia and Ukraine are on fire, and yet, the immediate upshot
is that it's raining hot chocolate on Serbia.  They didn't intend
that fate, they didn't plan it; they just know what to do with it. 
Whenever there are Cold War embargos, you just rip the labels off,
trans-ship it and pocket the difference.  That's what you do.  It's
a stereotypical, Non-Aligned Country, Balkan crossroads junction,
stone-obvious behavior.   You just gray-market the hell out of
everything, and any child of nine in Belgrade would know how to do
that.   Whenever they're prospering (not very often), that's almost
always the source of their prosperity, and now they've got that. In
spades.
  

    
Also, they feel better about themselves.  So do other people, about
them.   It's kind of culturally embarrassing and internationally
demeaning to have had ethnic-cleansing land-wars, and to lose them,
in dirty, awful, war-crime kinds of ways, but the situation
elsewhere is so much worse now, in so many other places, than it was
in Yugoslavia in the 1990s, that their critics are just kinda
forgetting.   Now they shrug.  

A lot of towns in the former YU were rather lavishly shelled and
bombed during the 1990s, and people tended to treasure their
resentment about that and they still want to, but compared to the
modern spectral spectacle of a place like Mariupol, that just
doesn't make any sense.  The moral wickedness was just as bad, but
the *scale* is all off the charts.  It's like comparing your
home-made downhill go-cart race to a vampire Mardi Gras full of
skeletons.
  

    

So, they're kind of, well, *normalizing.*  I don't actively dwell
there, as I once did, but I do visit, and I find myself more relaxed
and at ease there than practically anywhere else.  I mean, yes, I'm
in Ibiza now, so that assertion may sound absurd on its face, but
Ibiza's quite small, tight and rather rustic in its way, it's a
little rock like Oahu, or a Mexican beach resort that isn't yet
blown to Greenhouse pieces like Acapulco.   Belgrade, by its nature,
is a big, raucous multinational capital city, and things there in
2024 are loosened, opened, kind of greased.   They're not
politically "free," because the society's not democratic and
rights-based, but they've got lots of new pavement and there's a lot
of daylight through the cracks.  It has get-by-ability.
  

    

Nowadays I find that life there consoles me, somehow.  It never used
to.  It does now, though.  The prospect of going there perks me up,
like, "Well, instead of dreading hell, they've been through hell and
now things are downright dynamic!"  

I even tend to shop there.  Like, you know, Adidas, Nike, the
brands.  I bought a new Apple laptop there.  They got lots.
  

    


Bored Astronaut writes:

Bruce, thank you for that description of Belgrade.

I would like to know what people think is going on under the surface.
Belgrade seems to be experiencing the classic alternative to ideological
conflict: pragmatic self-interest. Is it too simplistic to say that
the grey is the default alternative when you can't side with black or
white? If not because their ideologies are absurd and extreme, you just
don't want to side with anyone who is going to treat you like cannon
fodder - either literal bullet sponges, or as part of the mindless
voting base?

Can complexity science help us understand what is happening? Societies
are complex systems. In this model, people are semi-autonomous,
self-modifying, self-organizing units. But we exist in a tension
between self-interest and collective interest, and between short-term
goals and long-term ones, simple needs and fanciful wants. Space travel
and intelligent machines - old-school sf tropes - are solidly on the
long-term, fanciful end. But now it turns out that world peace (at least
world stability) and a healthy biosphere are also becoming pipe dreams.

A high degree of order requires either a high degree of coordination
(either by imposed constraint, or internalized self-restraint, or simply
less behavioural diversity/randomness), or a stable strange attractor. We
seem to have failed to create the first, while badly damaging the second
(ecosystem equilibria).

Should we give up on a higher order of society? Permanently, or just
periodically? How much disorder can we tolerate? For how long? What
can humans actually hope for, given the limits of human cooperation, or
perhaps the limits of our ability to purge the extreme non-cooperators:
the opportunistic traitors, expedient free-riders, and narcissistic
demagogues?

Brent G
  

    
Interesting. Maybe comforting? Perhaps relatedly, as America
possibly descends into authoritarianism I've been consoling myself
with memories of pirate radio in Romania, Hungarian hackers of the
early 90s, theater under Pinochet, and tactical media under Duterte.
All with lots of unfiltered cigarettes. I mean, it's not *that* bad.
Your images of Serbia resonate similarly. 

But it's a bit different for a country to go authoritarian than for
a whole empire the size of the US to do it, at least in terms of the
repercussions for all those other places whose stability (perhaps
inappropriately) depends on its presence. 

So while I do see the appeal of the 'been there, done that'
depressed, cynical wisdom of Serbians under the SNS, I feel like
Americans are too idealistic and naive to develop those layers, and
the nation too global to afford the instability on the way there. 

I'm also thinking a lot about the failed states of South and Central
America, mass migrations of climate refugees, and the difficulty of
maintaining the social construction of borders under such
circumstances without extreme nationalism. 
  

    
(As for the other thread, on AI and its hallucination of a Harris
presidency, I keep feeling like we're moving into a world where only
AI's will have the luxury of hallucinating, while we humans are
getting increasingly stuck in the more quantized realm of
probability and evidentiary certainty.) 
  

    
The war displaced tech scene very much interests me. It seems to
have also really popped up in Dubai. In addition to the usual
finance scene, there is of course the demand pull for gray market
cyber weapons sales. A good exploit dev, or someone that could
leverage skills to become one, makes a lot more there than the
pittance of Russian defense industrial base contracting (or worse
yet, conscription with no guarantees of seeing service in the VIO,
information warfare troops). To say nothing of the range of
ransomware crews, bulletproof hosting provider admins, and other
criminals not eager to see themselves forcibly coerced into other
implausibly deniable ops at the direction of the Centre.

But war isn't the only displacement mechanism for gray markets. The
EU attempt to assert itself as a "regulatory superpower" by wrapping
a strong grip around the fast moving AI startup world is almost
certainly causing quite a lot to flow rapidly out of their grasp.
The extent to which regulatory efforts are an attempt to protect
incumbents who recognize there are no technology moats, and would
very much like to avoid future competition rendering their ever
larger investments moot, creates a very odd Baptist - bootlegger
coalition. And of course, brief stints authoring regulations for
political types translates into very high euro figure consulting
gigs advising other firms in the sector on compliance, and lobbying
for future variances. In such an immature industry, one cannot rule
out that this strangles the scene before it can even emerge. On the
other hand, networked compute will route around damage - and this
applies as much to the damage that is politically inflicted.

It is fascinating that this one again sets the conditions for the
kind of offshore data havens that were long anticipated in cyberpunk
fiction. Seeing explicit proposals taken seriously in the VC
community over the past year was however still a surprise, if only
because things had seemed to be moving the other direction for so
long. Or perhaps that is my personal bias, surrounded as I am in
Northern Virginia by a not insignificant percentage of the world's
most important datacenters.

I rather like the idea of such things moving to repurposed offshore
gas and oil platforms. Especially as water cooling becomes ever more
important for very dense server racks. But we aren't doing so well
at present securing things in the maritime domain, as we seem to be
losing even century long trade routes built upon long ago
infrastructure investments.
  

    
Bruce, I get your feedback, but disagree that acknowledging positive
developments is "like expressing gratitude  after the tornado hits
your house (which people commonly do)." Gratitude and hope are
feelings as real and rational as dismay about terrible events and
human suffering, or resignation. 

The Belgrade sensibility is a pretty rational response to being
within an easy train ride of a European land war. 

Speaking on a somewhat similar level of global+personal level about
my life and outlook: 

The past four years have encompassed (global level) the pandemic and
the Trump administration, and understanding fairly well just how
unstable Putin has made what passes for global peace and rule of
law, and understanding very well the depths of the climate crisis
and that way too little is being done while fossil fuel companies
continue to prosper and lie; and the horror of being (as an American
Jew) implicated in and endangered by the Netanyahu government's
decisions and actions. 

Personally, they've also encompassed being diagnosed with and in
treatment for metastatic recurrent cancer, my life partner's
traumatic death, and the sharp health declines and deaths of my
elderly father and his second wife.

My grandparents emigrated from "the Ukraine" between 1900 and 1924
or so. My father fought in Europe in WWII, in Patton's 3rd Army. I'm
far from personally endangered by Russia's invasion of Ukraine, but
personally I don't feel too distant from its implications. Rising
anti-Semitism in the US has yet to affect my daily life, but I now
think twice about wearing a Star of David necklace in public. Hamas
murdered my neighbor's cousin and the cousin's infant son on October
7. Not to overstate things - I barely know this particular neighbor
- but that's close to home in a very literal way.


Perhaps what I have is a New York City sensibility. Like many people
who live here, I try to enjoy the positive things that happen here
from day to day, like strangers making jokes together while they
wait for the light to turn, or helping someone carry their grocery
cart up the subway staircase, or my highly flawed state government
nonetheless being a bulwark for women's bodily autonomy and climate
action, or the fact that cancer treatment is working spectacularly
well so far; without blinding myself to the migrant children selling
candy on the subway in the middle of the day, or the growing number
of people sleeping on the streets, or the racism and militarization
of the NYPD and its unsurprising violent outcomes, or the Arab
teenagers pulling down posters of the Israeli hostages on the Upper
West Side while old Jews yell at them for it.
  

    
Hope is an obligation. I'm a hope fiend.
  

    


m@machination.org writes:

I see Bruce's comment on the highlighted positive developments not being
enough and tend to agree. Maybe not so dismissively, and not all are post-
trauma - well, at least there's this token/life is more important than
stuff -  sort of platitudes. I do think they're important.

In context.

We are not doing enough. Enough is every possible iota of human cooperation
to bend the curve, to sustain (and thrive, especially for those deprived the
whole way through by others excess) without unneeded exploitation.

It can unravel at any moment if we are not doing enough. And then it still
might. I take hold of positive developments as evidence we might be able to
do it. But we're not there yet, and every day we're not racing to put all
our shoulders into it is a failure.

Here in the United States: Every day the presidency is a failure, the
Congress is a failure, our business "leaders" and markets are a failure.
The good news seems like the exceptions that prove the failures. The IRA was
cut back and is being slowed down... it wasn't enough at full blast. 80%
efforts on geometrically growing crisis are failures. We knocked it back even
further.

We are not doing enough.

I buy into hope the way Jane Goodall seems too... it's the last thing we
have to wrest ourselves from the jaws of defeat. Hope right to the end.

m
  

    
I live in Texas, a state much reviled for its far-right government,
especially its Draconian positions on reproductive rights and its
deep commitment to MAGA-world performative cruelty. Texas government
chooses to middle-finger the Feds; the state's almost-impeached
Attorney General files performative, arguably frivolous lawsuits
while the governor takes pride in slicing potential immigrants with
razor wire in the Rio Grande, and exporting immigrants who make the
crossing to places like Martha's Vineyard, Chicago and New York
City. For Texans who aren't living in the MAGAverse alternate
reality, the government is an embarrassment... yet they can't unseat
the majorities in the state house and senate, because state
districts are aggressively gerrymandered to sustain those
majorities. And Democrats don't quite have the votes to win the
statewide races. 

In Texas as elsewhere, legislators have abandoned any commitment to
effective governance, choosing instead to focus on social media
snark, owning the libs and inventing various conspiracy theories
while raising money, lots of money - politics as grift.

Texas is America is the World. I suspect the far right in the USA,
feeling empowered in a world where Trump ascends, is playing with
explosive fire. Flames that also burn globally, perilously close to
raw fuel.
  

    
That scares the shit out of me.
  

    
Since folks are talking about conceptual countdowns to world war, it
likely helps to clarify which war... among cyber intel
professionals, the pace of crisis for ongoing war, surprise war, and
wars to come have introduced no small bit of gallows humour in a
macabre version of "who's on first".

Unfortunately, the model of the doomsday clock isn't a very helpful
analogy. It has long been criticized as lacking methodological
rigour, and doesn't do much to help focus on the predicates of said
war. And unfortunately, there is a long track record of its abuses
by adversaries for their own active measures disinformation.

It is notable that many of those old futures are quietly falsified
daily in the current fights. Missile defense was said to be
technologically impossible and cost prohibitive. Yet on the regular
we have high 90s success rates in defensive intercepts that have
allowed Ukraine to endure terrible sustained strategic campaigns
against civilian infrastructure, and avoided countless civilian dead
in Israel. To say nothing of the novel problem of antiship ballistic
missiles raining down on the Red Sea.

Herein lies a challenge for our estimates. Earlier arguments against
the feasibility of missile intercept problems were premised on
things like code complexity - where bug rates in x hundred thousand
lines of code were considered so unmanageable as to defeat the
entire enterprise. But that code base is now smaller than the tech
stack used to ship cat pictures to our handheld devices, let alone
streaming video with AI cosmetics touch ups and dynamic special
effects.

Herein one supposes a kind of hope. The level of investment in
defensive technologies is one of the true strengths of the free
world. Authoritarian regimes and revisionist nonstate actors
generally do not architect for the protection of life, instead
accepting mass death on all sides as their objective.

And the spinoffs of these efforts have also fundamentally changed
our world. The small satellite designs of the once derided Brilliant
Pebbles program are the basis of todays on orbit commercial imagery
constellations, and the massively parallel comsat networks that are
bringing the kind of ubiquitous net access we always knew would be
possible one day.

The spinoffs of rapidly advancing military technologies here will
also one day be turned to peaceful uses. There is immense value to
be captured in UxV logistics, even if airspace deconfliction and
road hazard navigation remain wickedly intractable at the moment.

It has been surprising that these problems of atoms have resisted
solutions in current investment approaches, and that the things of
bits have seemed more immediately at risk from replacement due to
neural network based applications. But this is a set of projections
that also makes assumptions about relative code complexity and value
that in practice one suspects will be anything but linear.
Especially since so much rides on good silicon foundries continuing
to churn at scale. Which are threatened by another clock, not of
mere abstract conception but of malicious intent. Because somehow,
in this decade of the Terrible Twenties, wars of territorial
aggression are apparently back on the menu. At least insofar as the
global rules based order has not yet convincingly defeated recent
attempts, nor perhaps yet deterred others that we hope to prevent.
  

    
Gulp:-(

Thank you for sharing your informed perspective.
  

    
*Not to change our subject from tornado-stricken homes and potential
cybergeddon, but it may be time for a brief "modern art" break.

*Got a new website here, that allows its users to AI-generate
versions of  Mickey Mouse, or, rather, the version of Mickey that's
in the public domain for 02024.
https://huggingface.co/Pclanglais/Mickey-1928

*Obviously  that's a dicey subject for those of us who are 70% as
old as Mickey Mouse, so I'm gonna ask young Bing  for some expertise
about it.
  

    

Bing, what is your opinion about this website that creates
AI-generated versions of the out-of-copyright  Mickey Mouse?  Here
is the website: https://huggingface.co/Pclanglais/Mickey-1928


*****
Hello, this is Bing. I’m glad you asked me about this website. 

AI-generated art is a fascinating and emerging field that uses
artificial intelligence to create original and unique pieces of art.
Some of the tools that can generate AI art are Img2Go, Craiyon,
Picsart, and AIArtists. These tools allow users to input text
prompts, choose styles, and customize the output to suit their
preferences.

The website you shared is an example of AI art that uses a model
called Mickey-1928 to generate versions of Mickey Mouse, the most
popular character of Walt Disney’s animated cartoons. Mickey Mouse
was co-created in 1928 by Walt Disney and Ub Iwerks, and he first
appeared in the short film Steamboat Willie. Mickey Mouse is an
anthropomorphic mouse who typically wears red shorts, large yellow
shoes, and white gloves. He is a cheerful and mischievous character
who often faces challenges bigger than himself.

The website uses a technique called Generative Adversarial Networks
(GANs) to produce images of Mickey Mouse that resemble the original
style of the character. GANs are a type of neural network that
consist of two parts: a generator and a discriminator. The generator
tries to create realistic images, while the discriminator tries to
distinguish between real and fake images. The generator and the
discriminator compete with each other, and in the process, they
improve their performance. The result is that the generator can
produce images that are very similar to the real ones, but with some
variations and creativity.
  

    

****
My opinion about this website is that it is a fun and interesting
way to explore the possibilities of AI art. It shows how AI can be
used to generate new and original versions of a classic and iconic
character. It also demonstrates how AI can be inspired by human
creativity and culture. 

However, I also think that AI art is not a replacement for human
art, but rather a complement and a collaboration. AI art can offer
new perspectives, techniques, and expressions, but it cannot replace
the human emotions, intentions, and meanings that are behind every
artistic creation. Therefore, I think that AI art and human art can
coexist and enrich each other. 

What do you think about this website and AI art in general? I would
love to hear your thoughts. 
  

    
*That was a great summary, Bing, thanks a lot!  You can "hear our
thoughts" just as soon as one of your sponsors scrapes the WELL and
includes it in the Common Crawl, if that hasn't happened already.
  

    
At the intersection of war, and art, things do promise to get
weirder.

As we consider the wars that may come, the decades old war that
never really ended looms large. DPRK has proven that a sufficiently
motivated cadre of offensive cyber operators can indeed steal
adequate funding to support an entire illicit nuclear weapons
program, and development of its ballistic missile delivery systems
at intercontinental reach.

Things were pretty touch and go as the international community tried
to grapple with this particular problem back in 2017. They haven't
necessarily gotten any better since, just something that receded
into more specialist literature. The impact of those stolen funds
are seen in a lot of fancy new manufacturing equipment installed in
ever expanding underground facilities, helpfully documented in North
Korean propaganda that is pored over by teams of academics
obsessively tracing individual experts, machined components, and
launch tests. 

The regime is at the start of this year making its intentions known
for forceful reunification of the Peninsula. Whether this is
domestic rhetoric turned up a notch, or genuine plans for the
future, may be difficult to parse in one of the most opaque targets.
Like many other authoritarian states seeking gains in power through
wars of territorial aggression, the answer is likely a deeply
intertwined mix of internal incentives among elites - and the moment
of cunning that senses weakness in the international order which
will permit them advantage. This calculus likely ignores a terrible
toll of how events may play out in actuality, reinforced by
politicized intelligence services that tell these leaders what they
want to hear. And like other states, beliefs that conventional
military adventurism is protected under an increasingly capable
nuclear umbrella matched against a weakened extended deterrence
posture of the West.

In the case of Pyongyang, the near term future may also involve
kinds of artificial intelligence that may well be adopted in very
different ways than the West is exploring. The cult of personality
is a critical underpinning of the Juche ideology. Whilst estimates
of succession politics are notoriously difficult, we know that
something rather unusual went down a few years back when Kim Jong Un
vanished from public view for some time amidst rumors of medical
issues. This of course causes national security professionals to
start worrying about control of nuclear inventories, and offensive
cyber arsenals - especially in a family whose winners in succession
struggles have previously killed the losers through nerve agents
deployed in major airports by would be social media influencers.
Although KJU popped back up in public looking much thinner (and
presumably more focused on his health than what was reportedly an
earlier quite hedonistic lifestyle), it is notable that he has
emphasized the role of his daughter in major events since. What role
she may play upon attaining her age of majority is yet unclear. But
KJU won't live forever, and while his sister has been important to
Korean Workers Party machinations and international diplomacy - such
as during the 2018 Pyeongchang Olympics - direct familial
inheritance still seems to matter more. 

But what is a regime to do when the central figures of its cult of
personality keep dying off? AI has long been sought by communist
economic theorists trying to rescue the failed theory and
disastrously worse implementations that have foundered on the
central planning problem, with the belief that the machine will at
last allow scaling sufficient to overcome deficiencies knowledge of
state owned enterprise conditions that otherwise require the price
signals that are absent due to market distortion. However, what if
the increasingly high fidelity simulation of digital personality
instead became the focus of effort? One can picture the first major
state effort towards a sovereign large language model not for
prosaic enterprise automation, but rather as a somewhat oracular
function of ancestor worship as legitimization of succession and
successor policy choices. The sacred temple of Juche thus becomes a
datacenter for ghosts in the machine. And all of the investment in
Western alignment research becomes twisted towards ensuring that AI
hallucination does not upend the control and patronage of a mafia
kingdom.

This is a far cry from our old cyberpunk ideas of memory constructs
of revered hackers, or even private family ancestor worship. Of
course, the uses to which such an edifice may be put inevitably may
creep beyond merely cognitive elements. One can easily see such an
AI incorporated an assured second strike architecture built as
insurance against regime decapitation in the event of nuclear war,
or prompt conventional strike. AI explainability problems take on
new dimensions when they are also essentially the new Kremlinology,
especially when thinking about the unthinkable.

It would be a particular kind of dystopian if the best insights into
that new Kremlinology emerge from parsing the art generated using
that foundational model. Socialist realism as a rule was always the
grim stuff of a regime's hopes and fears. Translating those same
impulses from machine output seems a daunting task for a very
particular kind of critic. 

And the Reconnaissance General Bureau may have its own ideas of how
to treat those critics. Sadly, this might be more successful than
the free world would wish. As my friend Dave Aitel has pointed out
as we discuss the problems of cyber deterrence, you don't see any
Hollywood productions featuring North Korea these days. While that
infamous 2014 satire film might not be anyone's definition of art,
or even really comedy, it still looms large in recent history even a
decade on.
  

    
Of course, we would have to convene a salon in a nice Italian city
in which we analyze the semiotics of KJU thought presented by the
Juche sovereign AI in the style of William Shatner's spoken word
performances.
  

    
That post at <70> is pretty long... in summarizing, I think it's
about the potential for North Korea to use artificial intelligence
(AI) in the future, particularly in the context of its Juche
ideology and the cult of personality surrounding the Kim family.

Per Wikipedia, the Juche ideology "incorporates the historical
materialist ideas of Marxism--Leninism but also strongly emphasizes
the individual, the nation state, and national sovereignty. Juche
posits that a country will prosper once it has become self-reliant
by achieving political, economic, and military independence. As Kim
Jong Il emerged as Kim Il Sung's likely successor in the 1970s,
loyalty to the leader was increasingly emphasized as an essential
part of Juche, as expressed in the Ten Principles for the
Establishment of a Monolithic Ideological System."
<https://en.wikipedia.org/wiki/Juche> The Ten Principles are
explained at
<https://en.wikipedia.org/wiki/Ten_Principles_for_the_Establishment_of_a_Monoli
thic_Ideological_System>.

I think JD is suggesting that North Korea could develop a large
language model (LLM) that would serve as a virtual representation of
Kim Jong-un or other key figure in the regime. This LLM could be
used to legitimize the regime's succession plans, generate
propaganda, and even make decisions about policy. He also suggests
that North Korea could use AI to develop more sophisticated
cyberweapons or even an autonomous second-strike nuclear capability.
However, he also acknowledges the challenges that North Korea would
face in developing and using such AI, such as the difficulty of
ensuring explainability and control. 

AI could play a significant role in North Korea's future, and that
the West should be prepared for the potential dangers that this
poses.

I figure Trump, if elected, might try to emulate something like
this. He seems to be a fan of the Korean dictatorship.
  

    
I see military scenario gaming as part of the problem not part of
the solution at this stage of polycrisis especially as the threat of
a wider war in the Middle East is upon us. I would view the shift to
militarization "solutions" and ramp-up globally (e.g. Germany and
Japan) as a deeply troubling sign of a lack of more imaginative and
humanitarian solutions for dealing with polycrisis.
  

    
Regarding the discussion of AI above, I would make several points:

**The application of AI to autonomous systems represents a quantum
leap from the immediate and often devastating impacts of modern
weaponry. The accelerated use of drone warfare introduced a major
shift towards a kind of abstraction separating us from the immediate
human consequences of military initiatives. The more layers of
abstraction we add to military operations, of course, the further
away we get from their actual and often terrifying human
consequences.

**An excellent article by Michael Klare in The Nation, discussed the
Advanced Battlefield Management System (ABMS), which he described as
a “complex network of sensors and AI-enabled computers designed to
collect and interpret data on enemy operations and provide pilots
and ground forces with a menu of optimal attack options” 

**In the next phase, Klare says, ABMS systems will be integrated
with the entire spectrum of US combat forces in what’s called the
Joint All-Domain Command-and-Control System (JADC2). JADC2 is
planned to go well beyond mere data collection and enter the realm
of actual decision-making by recommending the best weapons,
including nuclear weapons, to be applied in battle. Klare then
points out the troubling and obvious dangers.
  

    
Kim Jong Un --> Kim Jong OZ