Democratizing Forgery

DeepFakes have entered the arena of public interest in the form of a battle over censorship. Websites, most notably PornHub, have found themselves in the midst of a no win shit storm as a result of taking the proactive step to disallow content that is digitally altered so as to look like it contains recognizable notable people. Their reasoning: it’s a form of non-consensual content. The individuals who’s likeness is being used ostensibly did not consent to having imagery made of them in this fashion.

On the surface this issue seems almost childish. The internet is in a tumult over getting naked videos of a few hot actresses, as if their inevitable sex tapes were not already on the way. Dig a little deeper however, and you may find the act of superimposing a neural-network representation of Scarlett Johanson’s face on to someone else’s body speaks volumes about human desire, and the effort to prevent it says just as much about the landscape of our political world.

It starts with a simple human instinct. Men see the face of an actress in a movie, an actress famous because of her good looks, and they get to thinking about what the rest of her looks like. When the imagination and a few cartoonish artist renderings fail to fill the basest urge to mate with attractive women, the nerds get to work. A computer sifts through thousands of images of the actress’ face, piecing together what she really looks like. Then an unrelated video with a different kind of actress is selected and the neural-net paints over the face with it’s own idea of what a face should look like. If done correctly the effect is convincing enough to bring euphoria to the dry masses.

Ironically, the desire to see more than just Scarlett Johanson’s face has led us down a very complicated and winding road leading full circle to placing just her face on anything but her body. The effect of the compilation is more desirable than it’s parts. It is the combination of both, the face and the context it is in, that make DeepFake’s desirable.

Highlighting this division brings up the question, what part of a DeepFake is non-consentual? All the source material is obtained legally. The faceless woman agreed to be filmed and was compensated. The images that teach the algorithm are freely available. The final output of the process is a likeness but not a replication. It’s a machine’s interpretation, not a camera’s capture of a real person. If the process were a little less sophisticated, no one would care who’s images were use to produce it.

It’s only now that the process is good enough, that the likeness is close enough, that there begins to be concern. That concern does not rise from the fear that people will be tricked in to thinking the videos are authentic. The concern rises from the fact that people don’t care. The videos boldly advertise themselves as DeepFakes, and all censorship has been directed towards the videos that represent themselves as such.

Instead it is because they fill the desire to place the likeness of a carefully maintained public personality in a sexual context. They show the people what they want in a manner that bridges the gap between fabrication and reality in such a way that their imaginations are satisfied. They are the truest form of porn, adulterating not only the visual image of a person, but the idea of them as well.

Imagine a world where DeepFakes are used for political purposes. Not to seriously convince people of something untrue, but to play with the idea of a person in a fabricated context. A politician could be placed in endless scenarios, lay victim to the imaginations of his enemies, and viewers at home would engage along side in the fantasy.

We already exist in a world that substitutes facts for repetitious talking points and suggestive impressions. DeepFakes will be the nuke in that arsenal soon enough.

That leaves one question: why the backlash? Why the push to bury these things? The answer is that nukes are no good is everyone has them. This technology is cumbersome and hard to use right now but it’s going to become freely available soon, and when it does the only thing to mitigate the damage is the censorship measures put in place today.

Advertisements

Sri-Lanka Bans Facebook

In the wake of violent riots in response to the death of one man, the government of Sri-Lanka has disabled connection to facebook from their country. A delegation from facebook is set to meet with government officials to tailor a censorship platform that would allow the website to return to the country. Amidst the chaos and ideological tensions, a pattern begins to emerge that ties Sri-Lanka’s problems, and their proposed solutions, to issues world round.

It did not all begin with the murder of Kumarasiri, as many news outlets have written. The history and tensions within the nation are complex and long lived. Religious and political divisions define much of life, and violence, between groups that have been been in conflict for decades. The end to the civil war by no means brought about a popular reunification.

Knowing this gives us context as to how the murder of one man, though little information one way or the other exists on the incident, could ignite such an energetic and destructive response. To say that the propaganda spread on facebook is the cause of, and solution to, this riot is to say that matches are the cause of and solution to forest fires. No other extenuating circumstances exist. All decent can be squashed by deplatforming.

And in a sense it is true. Turning off facebook and other social media slows the spread of social action. That’s not to say it stops it, people were having far bigger and better riots before computers. It is to say that removing this form of communication is pretty good at keeping people from doing things you don’t want them to do. When it comes to stopping racial and ethnic tensions from flaring up its an easy sell.

When it comes to a tyrannical government strengthening it’s hold on a populous, it’s already too late to disagree. That is the situation in Sri-Lanka. Framing this a benevolent state attempting to quell unrest between unreasonable religious parties is a perfect cover for an incompetent state reeling from a recent civil war attempting to exert it’s power. Tell me I’m wrong. Tell me the primary response of a state to sectarian violence should be tinkering with censorship of speech.

We are all in danger of sliding down this hill. The desperation to solve violence once and for all will be our undoing.

 

 

Never Again Isn’t Enough

In the wake of last Wednesday’s horrific and increasingly predictable tragedy, a group of kids from the affected school’s theater group got together to have the first real conversation about gun control we’ve had in a long time, and the results might surprise you.

It was then that they decided to petition for more thorough background checks. As Alfonso Calderon, a co-founder of Never Again, who was there that night, told me, “Nikolas Cruz, the shooter at my school, was reported to the police thirty-nine times.” He added, “We have to vote people out who have been paid for by the N.R.A. They’re allowing this to happen. They’re making it easier for people like Nick Cruz to acquire an AR-15.”

New Yorker

Never Again may have finally found the answer to school shootings: keep the guns out of the hands of school shooters. It’s not even hard to enact, a simple background check into the criminal history of any gun buyer would stop the flow of firearms to the wrong people. The only thing stopping this new “background check” idea is of course US congress, stocked with members paid into office by the NRA.

Let aside the fact that Florida State has a provision in it’s constitution permitting, but not requiring, counties to instate background checks and wait periods for non-private sales of firearms. Let aside the fact that Broward County, where the shooting happened, chose not to enforce this portion of law. This really feels like the fault of the US Congress, and particularly the republicans who support gun ownership. They should have done more.

Now assume we do finish sweeping away all opposition to the idea of gun control, what is there to actually be done about the problem of school shootings? How, specifically do we control guns? Is it even possible to try in a country where there are slightly more guns than people?

The horrifying thought, and perhaps the reason that every shooting is followed with a ritualistic chant of the phrase “gun control”, is that we are actually powerless to do anything about it. It’s more than possible, it’s painfully obvious, that we have no control over the chaos. We have no control over the mass murders. We have no control over the loss of life. So we cling to the one thing we might have control over, the guns, but even those will slip through our grasp.

Even with a complete consensus that the shootings need to be stopped, there is no law that could be enacted in this particular country to make a serious dent in gun ownership for decades. It would be like outlawing grandmother’s fine china: every household has some, it’s easy to store away in the basement, and no matter how hard you try one member of the family refuses to part with it.

We need to keep the dream of gun control alive though. It is the only chance we have at staying sane through the hell of a country killing itself from the inside. Gun control is the only false hope that things will get better, not worse.

 

 

How News is Generated.

On December 1st, Vox published an article titled

A game of naked tag at a Nazi death camp was a Polish filmmaker’s idea of art

It’s got all the right buzzwords: Naked, Nazi, Filmmaker’s idea of art. Most importantly, the title conveys a sense of immediacy to the event, like it just happened.

You have to read past the opening paragraphs and scroll past the free rebroadcast of the video so offensive it sparked the article, to get to the little tidbit of information that the film was screened two years ago.

The only reason anyone is able to even call this news is that the exact location of the shoot has been uncovered as a chamber in the Stutthof camp. Oh yeah, and the film as shot in ’99.

The film found equally poor reception in ’15 when it was displayed at a polish museum. AP reports that it was taken down from Krakow and Estonia. In Poland the display was placed in a separate room with a warning.

The artist’s page defends the work as contextualizing the horrors of the holocaust. A fairer interpretation is that of dancing on the graves of sixty five thousand people. Its no secret that art has been shitting on important matters for the later half of a century, and this is nothing particularly new. It was filmed  18 years ago. There’s not much outrage left to be had over this one.

The question then is, why is this story being picked up by the likes of BBC, VOX, the daily mail, ABC, and BuzzFeed?

The Times of Israel seems to be the one that broke the story, stating “the Organization of Holocaust Survivors in Israel, the Simon Wiesenthal Center and several other groups sent the request for clarification to President Andrzej Duda”. The OHSI sent a letter. That’s why we’re hearing about this for a second time around.

This story is not news today. It barely passed as news when it was published the first time and even then it was 16 years after the actual occurrence. What pushes this particular matter to the front of left leaning news is the ability to once again use the word Nazi.

There are other art exhibitions that have sparked controversy in their time. There are other videos by the same artist. This one, however, hits the right tone to be featured in media. It sings with flagrant disregard for past tragedy and invokes the possibility in the readers mind that the newfound active Nazi party in america is in fact out there dancing on the graves of dead Jews. It’s perfect.

And to further dispel any lingering hint of sincerity to the outrage over this video, note that articles from 2017 show the video while the ones from 2015 do not.

Oh, and Trump is expected t make an announcement about Israel soon.

 

AI: not a big problem after all.

Yesterday, insiders from numerous think tanks and advanced computing companies came together to announce to the world that everything they’ve heard about Artificial Intelligence (or “AI”) over the last few years has been false. “There are still no computers that can think unique thoughts on their own. It’s all techno mumbo-jumbo and marketing speak to convince investors to invest in one company or the next. In fact, you may have been part of the entire effort to make AI seem more real.,” said the spokesman for the group, Nerdy McSoontobejobless. “Chances are that you’re in on the act, but just don’t know it. If you’ve ever been asked to prove that you’re ‘not a robot’ by selecting squares that include street signs, you’re basically spoon-feeding an database algorithm what a sign is so that standard text-recognition software can figure out what the sign says.”

Eepok

Ok, I understand that this user’s comments were in part satirical, but oddly enough they express a notion that is growing in popularity.

Robotic surgery isn’t any better.

Artificial Intelligence and the Modern Productivity Paradox – Paywall

Without Humans, Artificial Intelligence Is Still Pretty Stupid

If you happen upon these and other similar headlines enough times in a short period, you might think that something has changed in our outlook on technology. “Oh, sorry guys. We were pretty sure we could build something amazing, but we just finished adding up the numbers and, turns out, it’s impossible to do anything more that what were currently doing with computers. Yep, that’s right. Seems we’ve hit a ceiling.”

This is of course in stark contrast to other headlines like:

Technology invading nearly all U.S. jobs

It gets to a place where the average Joe might wonder what the real turn out of all this technology is really going to be. On one side we’ve go people saying that we can just fix the economy with a few tariffs and on the other you have Stephen Hawking warning us to arm up for the impending eradication of life at the hands of computers. We seem to be lacking a good way of judging the reality of AI and automation.

So, lets break out my favorite analysis tool” Process vs. outcome. We are, in a sense, concerned with the outcome of AI, more exactly we’re concerned about the future outcome. We want to predict where it will take us. To make that prediction we have two ways of thinking.

Outcome thinking will attempt to gather information about our current state in the present, maybe some previous historical states, and extrapolate on that information. It’s pretty weak at this sort of thing because it relies on accurately measuring parts of the real world we don’t really have access to. We personally don’t know how many jobs have been taken over by computers. We can try to estimate based off what articles in the news say, but then were back to the issue of sifting through contradictory headlines.

Process thinking requires that we simply attempt to answer a series of questions. Have computers ever proven capable of performing the same task as a human? Does the capacity of a computer to perform a task increase or decrease with the amount of computing power available to it? Does computing power per dollar increase or decrease over time? Is leveraging automation profitable? Is there something functionally non-reproducible about the human mind?

These aren’t meant to be leading questions. The answers to them do matter in how we draw our conclusions. What is important here is that the questions do not attempt to measure actual events or information, but rather the mechanism behind the trends. We don’t need to measure progress directly, we just need to know that it is inevitable through natural market forces.

The Demand for PTSD

PTSD is a difficult topic to grasp. Every case is different, and over all our understanding of the human mind just isn’t that good. Recently there has been a trend, a result of the Iraq war most likely, towards rear-stationed troops demonstrating similar post war PTSD rates as front-liners. It would seem that the cause of this condition is is not only in major traumas involving death and killing. It is also present in the day to day operations of relatively safe bases.

This is not my speculation, but that of Sebastian Junger in his book “Tribe – Homecoming and reunion”. In the book he makes the case that separation from one’s fellow soldiers is in itself a significant trauma, and responsible for at least some of the cases of PTSD upon returning from war. This is because the bonds between soldiers becomes so strong that departure from a deployment feels like leaving a family behind. Or, more aptly in the metaphor of book, leaving one’s tribe.

The importance of this is that the very brotherhood the soldiers form, and is so damaging to leave, is the result of a conscious and concerted effort on the part of military’s training system. Such strong bonds are not only needed to improve teamwork and cooperation, they are needed to impel soldiers to kill.

Getting soldiers to actually kill has been notoriously difficult throughout history. Dave Grossman’s “On Killing” is a study of the tremendous resistance the average soldier has towards killing, even in the face of his own death. The only time soldiers seem to be liberated to kill without compunction in in the setting of an atrocity.  When the soldiers feel their enemy is sufficiently “Other” or not fully human, they lose their resistance to killing and fall in line with social pressure or orders to do so.

This othering tactic is well known and has been implemented to varying success throughout the history of warfare. Only now in the modern era has is been perfected, not through propaganda and hate for the enemy, but through team building and bonding with one’s fellow soldiers. The reasoning does like this: The stronger the bonds between the soldiers, the more “other” all outsiders appear.

It’s wildly effective and todays troops have successfully been conditioned to execute their jobs, killing if need be, without the same turmoil as previous generations. The downside to this camaraderie is that it isolates the soldiers utterly. Not only does the enemy becomes more distant and less human, but friends and family back home do so as well.

A return back home is not a return to normalcy for the conditioned soldier. It is being torn away from the only world he knows and thrown back into an alien planet with people and places he is only partially bonded with. This creates the massive culture shock of returning from deployment, and in turn the high incidence of PTSD.

TO further this problem, returning soldiers find that there is no analogue for the camaraderie of the military. They leave a highly concentrated form of tribalism and loyalty to return to a disparate world of individualism. No longer possessing the tools, or perhaps having been given a taste of something more satisfying, they struggle to reintegrate.

The military can’t and wont give up this form of conditioning for its soldiers. THe civilian world will continue to be an increasingly individualist place. Thus, PTSD is an endemic and inexorable part of modern warfare.

2020 Election Predictions

It’s time to get the ball rolling on some 2020 predictions because I see the outcome of that election as being a major factor in the trajectory of this nation’s culture. It’s not because I believe the office of the president holds that much importance, rather because the election is the largest social experiment in the world. This past cycle we successfully brainwashed half the nation into believing we were reliving 1940’s German history, and the other half into thinking the host of Miss-America had the secret fix for a failing economy. The president is unimportant; the stories told about the president mean everything.

The teeter totter of Dem/Rep victory is the main mechanic of altering our social narrative. What I mean by social narrative is the set of ideas that are acceptable to promote and repeat. In our current cultural narrative is acceptable to promote and repeat that women suffer discrimination. It is not acceptable to suggest that our borders should be secured. Sure there are people saying it, the president is saying it, but the narrative has made it controversial.

Narrative is awarded to the party that can claim the greatest victim-hood. With the recent Rep. win, the conservatives are placated and the Liberals are in overdrive. The cultural narrative is predominantly a liberal one now because they are motivated to retake power and fight injustice. The conservatives basically have what they want and can only muster a few passing laughs at the fever of the other side.

Should the 2020 go to the democrats, the Republicans may regain their initiative and take back the cultural narrative, but only after the fact. They’re going to be sluggish compared to the left. Should it go to the republicans however, the left will essentially lock down their ownership of cultural narrative for the foreseeable future. It’s this dynamic that has me interested in the outcome of the next election cycle. The shape of this country rests on it.

There is little point in pondering the republican candidate. I see almost no chance, barring immense stupidity, that the RNC would make a serious effort to run someone other than Trump. If they did, it would be a huge risk based largely on a sense that Trump has become fatiguing. The left puts up so much resistance that even his supporters don’t want to hear about him anymore. Still, they would need to hit it out of the park with the new guy and I just don’t think they have anyone in the wings for that. Trump it is, then.

The Democratic candidate is the real area of interest. who that might be boils down to whether the DNC learned anything from their defeat in 2016. Based on their new slogan “A Better Deal: Better Jobs, Better Wages, Better Future.” It would seem they haven’t. They’re still clunky, awkward, and relying on their grasp of the social narrative to pull them through. They may still think their approach with Hillary was good and only failed because of exigent circumstances.

If they did learn from their mistakes, first round pick would be Joe Biden. He’s well liked, had a solid resume, doesn’t hold the same negativity as Hillary but can still project the message “Back to business as usual”. He would be seen as a return to the center, which is dearly needed. In a race between Biden and Trump, I think the vote might go to Biden. He could project  the sensibility and stability the swing voters desperately craved in 2016.

Problem being, he’s a white male. While adding no points for the swing voters, this absolutely deducts points for the core democratic base. The last white democrat to make it to the general was Kerry. Think about that. Somewhere along the way, the party made the decision to run the firsts game. First black president, first woman president, etc. Biden would be seen as a step backwards. They’d support him, but not with the same energy.

The alternative, the Hillary-should-have-won approach, is to bring in an outsider. I don’t have an actual pick for this option, and in talking with the politically informed, don’t think this person exists in real life, but he certainly was a character in the last season of West-Wing. Matt Santos, the vaguely Hispanic representative from Texas that ran and won after the end of the Bartlett administration. He’s a ready pick because he gets to continue the pattern of minority candidates but doesn’t shove anything down down our throats. Think of him like a Hispanic, probably not Mexican, Obama.

He could remain centrist on most of the divisive stuff while allowing his presence to signal social justice. His base could sling shit for him all day without his needed to say a word. He’d look clean to the opposition while still energizing the fanaticism of a culture war. Between Santos and Trump, I’m honestly not sure who’d win. And there in lays the danger.

A republican win proves to the fanatical left that they are in Nazi Germany after all. There will be no getting them back after that. Given their individualist/separatist leanings, it would drive the wedge deeper into the heart of this country.

A democratic win would leave the right feeling like they’ve lost their rightful turn. We’re used to two term presidents. To loose that on the grounds of the left shouting about social justice would leave a welt. Given the already high tension, it might be enough for the right to shake off it’s lethargy. A politically motivated conservative base is not something this country has seen in decades and isn’t something we’re ready to stomach.

The teeter totter would devolve into a tug-o-war. Balance would be traded in for domination. We as a country, unable to reconcile our differences, would only in-fight further. Without a centrist Democrat in office, there is no hope for america.

 

 

Projecting Fragility

It’s finally happening. Trumps wave of hate crimes are rolling in starting with the ban of heroic transgenders from the military. Only problem is that Trump’s base still doesn’t seem to see the problem. The disagreement on the nature of the problem is actually more interesting than the problem it’s self. Let me explain.

Trans have, are, and will continue to serve in our military to some extent and in some fashion. They were not allowed before, and still managed to find ways to serve, and may be banned again, and still find ways to serve under the radar.

What is being ruled out is transgenders openly serving. This is not like the Tuskegee Airmen being allowed to fly despite the color of their skin. A ban on transgender does not actually bar them from serving. It bars certain types of behavior. The whole argument is over behavior, not the nature of transgender as a life.

The pro-argument goes like this: Trans people are people and should be afforded all the same opportunities as everyone else. One such opportunity is to serve in the military. Under the ban, a dedicated soldier could be dishonorably discharged or even court-marshaled just for the fact that they identify differently. That is discrimination

The con-argument goes like this: Anyone who can be a combat effective soldier should be allowed, but trans people have major complications that routinely distract and hinder combat effectiveness. They are fragile, possibly mentally ill, and complicate every rule set in place. If they really want to serve so bad, they can suck it up and pretend to be normal.

The key to all this is that transgender is an identity. Being a soldier is in part about dropping one’s identity and becoming a piece in the greater whole. If trans feel that giving up an open expression of their identity is enough to keep them out of the military, then they should not be in the military. Even if that identity is deeply rooted, it’s not a part of the soldiers’ job to express it. Having sex is another deep;y rooted behavior that the military does not permit while on the clock.

 

Is Trump Fixing the Environment?

Just a small follow up on my previous post about the Paris Accord: Trumps initiative to withdraw from a meaningless agreement has lead a number of large US companies to commit to greener operations.

AppleAmazon, Facebook, Google, Tesla, Lyft, and Uber are among the U.S. companies that have added their names to the “We Are Still In” campaign

In the absence of leadership from Washington, states, cities, colleges and universities and businesses representing a [sizable] percentage of the U.S. economy will pursue ambitious climate goals, working together to take forceful action and to ensure that the U.S. remains a global leader in reducing emissions

http://fortune.com/2017/06/05/paris-agreement-apple-amazon-google-facebook/

Just as was pointed out a number of times, the effects of the Paris accord were voluntary for the most part. Anyone who wanted to could still abide by it’s guidelines without having to shell out $100 Billion annually.

What is most interesting about this is that in a perverse way, Trump has managed to spark a green revolution in US Industry. The hatred everyone feels for Trump has lead to them becoming better businesses out of spite.

The question is: Is Trump fixing the environment? Is his strategy of rolling back Washington’s control over the matter impelling private companies and citizens to do their part? If so, was it intentional? We already know just talking about immigration during the campaign had a sizable effect on illegal immigration from mexico. Trumps words might be stronger than his actions.

In the same vein, Trump has announced that Apple is opening three new plants in the US

http://www.cnbc.com/2017/07/25/trump-interview-apple-to-create-u-s-jobs-with-new-factories.html

I spoke to [Mr. Cook], he’s promised me three big plants — big, big, big

Notice how that doesn’t sound like Tim Cook announcing three new plants. It’s Trump pigeon holing Cook into doing it. I see three options:

  • Apple makes the plants
  • Cook gives trump the middle finger because that’s whats expected of him, even if his original plan was for the plants
  • Apple commits to the the plants and then quietly never builds them

That last one seems most likely to me. It’s still a win for us though because Apple committing to US production sends a big message that America is open for business.