No Way Forward, No Way Back

No Kings, Russia / Ukraine, and More On AI

Dirk Diggler Explains World History to Donald Trump

From the movie Boogie Nights:

Dirk: Napoleon when he was the king, you know, people would just constantly try to conquer him, you know, in the Roman Empire, so it’s ...

The No Kings March

Preface

For those who may have forgotten, the (Barack) Obama administration was warned when it shifted the judiciary function for capital cases abroad to the White House that doing so would come back to bite the Democrats. While Democrats trusted Mr. Obama to adjudicate fairly and only kill (as a King would) those who were deserving, few others in the world did. To now complain that too much power is concentrated in the Executive Branch would be rich if Democrats had any knowledge of what I am referring to.

It is the Democrat’s inability to self-reflect--- a product of their near-complete ignorance of the policies that they claim to support, that makes them so repellent to so many. While the following was as true of George W. Bush’s supporters as it is with today’s Democrats, those most supportive of the party know the least about its actual policies. Barack Obama’s economic policies, in particular his bailouts of Wall Street, were amongst the most socially destructive acts in modern American history.

To get there, I must first address the destructive nature of the US Party system to get out of its vortex. From the time that I was legally eligible to vote through 2008, I was a registered Democrat. I voted for Barack Obama in 2008, so happy to be rid of the George W. Bush administration that I could barely sleep on election night. Three months after Mr. Obama’s election in 2008, I swore that I would never vote for another Democrat. While I’ve never voted for a Republican, I am quite certain that I will never vote for another Democrat.

Righteous anger at the Democrats over Barack Obama’s Wall Street bailouts led to the Democrats losing control of Congress in 2010, and with it, the power to get their legislation enacted. Mr. Obama went on to oversee the evisceration of the Democratic Party at the local and state levels. Those attributing the loss to racism missed that Mr. Obama’s policies had driven tens of millions of people from their homes through foreclosure as the rich were being handed billions in free money by the Obama administration.

Confusion over the nature of the bailouts amongst Democrats persists. Few argued at the time that nothing should have been done regarding the self-devastated banking sector. But banking crises have been around for as long as banks have been. Prior to 2008, a protocol had been established whereby banking systems were resurrected with Federal help, but culpable bankers were sent to clean toilets in public schools or some such, not handed billions of dollars in free money to squander.

Back in the day, there were many good suggestions for what the Obama administration should do about the crisis. There was the ‘Swedish model’ whereby failing bank assets were separated into good and bad. ‘Good’ banks were created to host the viable assets while bad assets were sold and culpable bankers were fired. I recommended (to financiers) either the nationalization of failing banks to then resolve them or the Swedish model. But as Mr. Obama’s Treasury Secretary Timothy Geithner put it, ‘America doesn’t do nationalization.’

Although Geithner’s was weak history, fair enough, so what was the plan? What the Obama administration did under the ruse of ‘saving the system’ was to hand billions in Federal largess to Mr. Obama’s friends and donors on Wall Street while he used trillions in Federal largess to rebuild the financial system that had just killed the world. This makes the Democrats now complaining about oligarchy almost too stupid to countenance. Oligarchy in the US would have disappeared by 2010 had Barack Obama not made it his life’s work to revive it.

This isn’t to blame Mr. Obama for his time and place. Following from Bill Clinton, the Democrats are the Party of Wall Street, the rich, and ‘liberal’ class warfare. Even given the evidence, gullible Democrats still imagine that the ACA (Obamacare) was about public health rather than making Mr. Obama’s donors in the health insurance industry richer. And the automaker bailouts codified ‘tiered’ wages, assuring low and declining wages for younger workers in the US auto industry. Mr. Obama was no friend to labor.

Quite remarkably, today’s Democrats make no association between Barack Obama’s policies and oligarchic wealth, in particular, Donald Trump’s. Had Mr. Obama focused on saving the rest of the country rather than just Wall Street 2009 – 2016, the US likely wouldn’t be at war with Russia in Ukraine, in free fall in terms of international relations, some utterly useless Democrat would be in the White House, and the economy would suck a whole lot less in terms of how the American people get by.

Had a bunch of us not predicted that this is exactly how Mr. Obama’s policies would turn out--- by early 2009, naysayers might have a point. Recall that it was the DNC that decided to call those of us who objected to Mr. Obama’s policies ‘racist.’ This became the Democrat’s mantra, lever, wedge. Recall that I voted for Mr. Obama in 2008. My objection to his policies had nothing to do with Mr. Obama’s race. The objection was that Mr. Obama was a neoliberal POS who launched amongst the most vicious class wars in modern history, with full backing from American liberals.

End of Preface

For those who may have forgotten, the last US President from the Democratic Party was a xenophobic, mass incarceration loving, proponent of every crappy American war of the last half-century, who instigated a war of choice in Ukraine while he participated in an ethno-nationalist genocide in Gaza. In terms of domestic policies, he promised quite a bit in order to steal the nomination from Bernie Sanders, and then failed to deliver on almost all of it.

To the ‘we would be at brunch’ signage of the No Kings participants, words fail. Your boy (Biden) started an honest-to-God WWII style genocide in Gaza as he goaded Russia to attack the US with nuclear weapons. Had the signs read ‘we would all be dead now,’ at least some connection to planet earth would have been in evidence.

Tell me, where do the national Democrats stand on the genocide in Gaza? Why not ask Kamala Harris? Her position as of 2024 was that ‘Israel has a right to defend itself.’ What makes this response so vile is that Israel isn’t defending itself in Gaza. Under international law and the judgment of the United Nations, Gaza is occupied territory, giving Gazans (including Hamas) the right to defend themselves from the Israelis.

Under international law, the Israelis have no rights in Gaza. This is because Gaza isn’t Israel. Israel is an occupying force in Gaza. And frankly, I don’t know many adults who buy a word of the official story out of Israel regarding October 7. Tellingly, the Israelis haven’t conducted a formal investigation into what happened. The early facts that did come out of Israel were not supportive of Israel’s explanation of October 7.

Under Israel’s ‘Hannibal Directive,’ half or more of the Israelis killed on October 7, 2023 are thought to have been killed by the IDF (Israeli soldiers). According to The Grayzone’s Max Blumenthal--- who did a deep dive into the evidence, there were no ‘babies baked in ovens.’ There were no mass rapes of Israeli women and girls. Almost all of the extreme violence seen that day was carried out by the Israelis using weapons supplied them by the Americans.

Now recall Blue Team dotard Biden swearing that he had seen pictures of ‘babies baked in ovens’ up to a dozen times after his own staff had asserted that ‘the President misspoke.’ This would be one thing if Biden were simply lying about his four decades of trying to cut Social Security and Medicare. But he was lying to support a genocide that he personally thought was a great idea.

As Biden was leaving office the Israelis were asked how many times Mr. Biden had petitioned them to end or limit the genocide. Their answer? None. Never. Not once. But that isn’t what he told his constituents. Recall, had Biden not geezered-out during his debate with Mr. Trump, the Dems never would have turned on him. The implication: the Dems are fine with genocide. They love it. It is their favorite thing. Even with the evidence of genocide in hand, the national Dems are still as enthusiastic about the genocide in Gaza as the Israelis are.

On the racism front, and apparently unbeknownst to Democrats, Donald Trump is just now approaching Joe Biden’s record of crafting and enforcing racist policies as a way of slandering and politically repressing poor people. Biden’s reputation as ‘Cracker Joe’ was the reason why Barack Obama chose him as his Vice-Presidential pick--- to draw racist Democrats to his campaign. On the rare occasions when Mr. Biden wasn’t opposing racial desegregation, he was a confederate of Southern segregationists Strom Thurmond and Jesse Helms.

Mr. Biden’s reputation for expertise in foreign policy came from his helping George W. Bush sell his war against Iraq to reluctant Democrats. One-million dead Iraqis and West Asia set on fire later, Mr. Biden simply lied to assert that he had opposed the war all along. As in his last debate with Bernie Sanders, American Democrats heard a man who had spent his prior forty years publicly trying to cut Social Security, Medicare, Medicaid, and Veteran’s Benefit, state that he had never done so. And this jackass was going to ‘save’ Social Security? Raise your standards people.

Update on Russia / Ukraine

Note to readers: this section was substantially completed (since rewritten) when news came that the upcoming meeting between Vladimir Putin and Donald Trump in Budapest has been cancelled. Russian Foreign Minister Sergey Lavrov and US Secretary of State Marco Rubio held an extended telephone call in anticipation of the meeting. That there was no clear path forward was the source of the impasse according to Mr. Lavrov.

How not constructive the American position is in a geostrategic sense is difficult to put into words. From the American perspective, the war in Ukraine is a plaything, a toy by which American leaders imagine themselves to be both brave and competent from the comfort of a warm chair buried deep in the bowels of some protected location. In contrast, the Russians will either prevail and the West retreats or they will be perceived as a future victim of American imperialism. Since 1917 at least, US Presidents have been lusting after Russia’s resources.

If memory serves, keeping Western hordes out of Russia is the primary purpose of the Russian Nuclear Doctrine. What the Americans haven’t yet figured out is how to get around this doctrine in their pursuit of filthy lucre. Readers may recall my hypothetical of robbing a mini-mart or the Post Office of $100. In the first case, if caught, the robber goes to prison for three years. In the latter case, the robber goes to prison for 20 years. For the same $100. Following three years of war, the Americans still want to rob the Post Office.

Here, current events in Venezuela come to bear. Venezuela has the largest proven oil reserves in the world. The US is using the easily disproved pretext that Venezuela is involved in narcotic production and distribution to gin up a war. So, Americans can believe lies from the Trump administration regarding drugs, or they can look at the last century’s history of US resource imperialism to see the real reason why Venezuela is a target.

How this relates to Russia in the present is straightforward. From Woodrow Wilson to George W. Bush, Americans have been lusting after Russia’s vast resource wealth. The US spent its time from 1991, with the dissolution of the USSR, to 2022, the launch of Russia’s SMO, provoking Russia into a fight that the US was certain that Russia would lose. With the US not having ‘won’ a war since WWII, the loss in Ukraine should not have come as a surprise. And it was the Russians who won WWII in the European theater.

The frivolousness of the American position--- as a late imperialist power that never wins the wars it starts, but that also never pays a military price for losing them, lies in the war in Ukraine being a vanity project by and for American liberals. The unity of the Biden administration, the DNC, and the CIA allying to launch a vanity war for the benefit of Western economic interests is a far cry from the reasons that we are given for spending $1 trillion per year on ‘defense.’

The Trump administration’s renamed Department of War is a return to Teddy Roosevelt’s Rough Riders, which US General Smedley Butler later described as being a ‘gangster for capitalism.’ The context was General Butler invading and looting other nations in much the same way that Donald Trump is currently eyeing Venezuela. During George W. Bush’s catastrophe in Iraq, the imperialist phrasing changed to ‘kick their ass, and steal their gas.’

The US started the war in Ukraine via the coup in 2014 and subsequent invasion by the CIA (and here). The CIA built the ‘Ukrainian’ army to attack Russia with. The CIA surrounded Western Russia with US spy equipment, a proxy US army, and then sent Ukrainian fascists to slaughter Russian speakers in Ukraine’s East in order to provoke a conflict. The US then invented the propaganda talking point of ‘an unprovoked, full scale, invasion’ to cover its tracks.

Back to the future, the Biden administration, the CIA, and the Generals screwed the proverbial pooch when they arrogantly launched a war of choice against Russia that they then lost. The situation where subsequent administrations are left to clean up the mess is also the one where they are expected to take the fall for the CIA’s poor strategic reasoning. The result: the US now has no way to win, and no way to lose, the war with Russia. And Donald Trump is beginning to realize this truth.

While the Trump administration is holding open the possibility of future meetings with President Putin, what is apparent is that Mr. Trump has little idea of what he is doing. The Russians have substantially won the war in Ukraine, but Mr. Trump is asking them for concessions. We not-Donald Trumps concluded in a matter of seconds that the Russians will significantly set the terms for ending the conflict. In this light, the Trump administration has concluded that not enough Ukrainians have died yet. This conclusion is on the US, not the Russians.

Moron (More On) AI

Last week I provided an intuitive explanation of why Google AI reproduces the dominant narrative on contentious issues in query results. This week I provide an (intuitive) overview of the mathematics, because they make AI the absolute wrong choice for many of the applications that it is being applied to. As it turns out, the AI process is all math, statistics mainly. AI uses the (augmented) probability distributions of the words, phrases, and the order in which they appear, in the training sets from which AI draws its ‘knowledge.’

In plain language, the more times that words and phrases appear in a certain order in the training set, the more heavily AI will weight them in its responses. Down a bit in this essay readers can find the text of AI ‘explaining’ how this process promotes dominant (state sponsored) narratives. Briefly, on political and geopolitical questions, state propaganda provides content for the establishment media. And the establishment media then magnifies the state propaganda.

It is easy enough to see in this formulation that state propaganda will be the ‘go to’ response of AI due to the larger number of occurrences that this process of magnification by the establishment media produces. In other words, there is no ‘intention’ by AI to promote state propaganda in its responses. That it does so is a function of the predominance of the state--- either directly or through this process of magnification, as the source of political and geopolitical reporting in the US.

That this wasn’t understood in the development of AI suggests that coders and mathematicians be compelled to study Marxist economics. Without getting too technical, the idiocy of the Clinton era ‘chaos theory’ movement in mathematics combined with the ‘wisdom of crowds’ fantasies of the (finance-employed) libertarian-Right led to an uninquisitive view of statistics held by AI developers. Question to statisticians: how were the distributions of the underlying variables decided upon?

Rephrased, why is the statistical premise implied that words, phrases, and their order are randomly distributed rather than clustered around sources possessing economic and political power? As readers will see below, Google AI provides an excellent and reasonably detailed explanation as to why doing so simply restates the power relations of Western societies in statistical form. To be clear, it isn’t even a consensus view that AI produces. It is the view from power that AI delivers.

Readers who imagine that mathematics is ontologically neutral can read my detailed explanation of why this is not the case here. Some years back I had a discussion of this problem with a mathematician who was working on what would become AI math. He offered that the one thing that mathematics is not good at / for is counting. The problem is in defining what it is that is being counted (see my book for details). Probability measures the number of occurrences of X given total occurrences. AI is a counting exercise.

This puts a different, and more politically repressive, light on the American practice of censoring dissenting views. Again, think of this in terms of how many times censored views are likely to appear in a training set versus the views of the state and state media. One set of views is suppressed while another set of views is magnified. And AI was engineered to count the number of occurrences of each and weight them accordingly.

Graph: the graph represents a conceptual picture of how AI source weighting works. In late 2002, your author concluded that the official narrative around Iraqi WMDs was cynical bullshit. Meanwhile, the New York Times was promoting the fabrication that Iraq possessed WMDs. As history has it, I was right and the New York Times was wrong. However, given that the New York Times’ reporting was 1) state propaganda that 2) a New York Times reporter (Judith Miller) reported as fact, 3) that was then magnified globally, AI would have given 1,000X the weighting to the New York Times that it gave to the Journal of Belligerent Pontification. Source: Urie.

In terms of AI being naively brilliant, this refers to its deference to social logic rather than possessing analytical methods that it then applies to the underlying questions of interest on its own. Philosophically inclined readers will recognize the reach of philosophical Postmodernism here. The realm of AI is social, not physical. AI ‘trains’ on texts that reflect human interpretation of facts, not on the underlying facts themselves. This is, in fact, an implied restatement of the postmodern conceit that scientific knowledge is ‘socially constructed.’

But the suggestion here isn’t that a cabal of postmodernists applied their theories to AI. The bet is that the creators of AI never heard of either philosophical postmodernism or even philosophy, except for formal logic. In philosophy of science terms, formal logic, like its mathematical sistren, depends on infinite regress as its ‘proof,’ meaning that in scientific terms, the methods of science don’t meet the minimal criterion that science be premised in demonstrable facts.

By deferring to the scientific literature instead of conducting science itself, AI was removed from one realm of contention--- the question of what it is that science actually does, to the question of how to compile scientific inquiry without possessing the ability to distinguish analytically likely stories from unlikely stories. Again, at its core AI is a counting machine not unlike the pressurized hoses that are run across roads to count the number of passing cars.

While few people would likely call these hoses ‘intelligent,’ they produce data that is then abstracted to the point where the fact of why they aren’t intelligent (they are pressurized hoses) isn’t immediately obvious. So, rather than laying the pressurized hoses on roads and doing the counting, AI takes counts processed from the work of others and surveys the social responses, weights them, and then compiles what is intended to be narratives around them.

For fans of economic history, probability theory came to prominence through its industrial applications. Physical production processes (a/k/a making stuff) have quantifiable characteristics like an error rate that can be calculated by dividing the number of errors by the total number of products produced. This can in turn be used to make predictions. A 1:10 error rate means that for every ten items produced, one will be defective and nine won’t be.

A necessary premise of using a probability-based approach to AI would be that the subject matter of interest be randomly distributed. (Because otherwise there are factors driving the data that aren’t considered in modeling it). This premise would emerge from the libertarian / neoclassical economist’s premise that concentrated social / political / economic power in the West does not exist. Yes, borrowed from educationally lobotomized economists, Jeff Bezos and my neighbor who works the midnight shift at Dunkin Donuts possess the same quantum of economic power.

With a little work from yours truly doing the prompting, AI offers (below) a very good explanation of why the premise that concentrated social / economic / political power does not exist is refuted by AI results. In fact, AI results refute the base premises of 1) democracy, 2) capitalism, and 3) the statistical distributions used in the social sciences. The mathematical approach taken by AI developers represents the Western liberal ethos that we are all equally endowed actors, meaning that truth is determined in the ‘marketplace of ideas.’

That the AI results below destroy this fantasy points to the paradox that while AI isn’t a thinking machine because it doesn’t think, it counts, and counting requires thinking to be chosen as the method of AI. Further, if AI is conscious, or at risk of becoming conscious, so is a pocket calculator. Ironically, with a few well-phrased questions, AI will tell you that it isn’t what the people selling it claim it to be. Not even close.

Two ‘conversations’ that I had recently with Google AI are provided below. These responses appeared toward the end of extended back-and-forth over the methods that AI uses, as well as the implausibility that they will ever produce political information that isn’t state propaganda.

Regarding interpretation, a note of caution. This is not a ‘gotcha’ regarding AI. These problems have been well understood for a long time. They are ‘understood’ by AI as well. And note the AI’s jab (below) at ‘monetization’ as driving AI creators to sell what they know to be an unsuitable product. There are applications for which AI could be useful. But by selling it as that which it is not, an enthusiastic push back is on the way.

AI ‘Conversation’ 1:

Question: (D)o you understand why (AI) answers that reinforce the dominant narrative are political?

Answer: Yes, the reinforcement of dominant narratives by AI answers is fundamentally a political act. It is not an inherent or neutral function of AI, but a direct consequence of the political nature of the data it is trained on and the decisions made during its development.

Here’s a breakdown of why this is the case:

1. Training data is a political artifact

The vast datasets used to train Large Language Models (LLMs) are not neutral collections of information. The content is created by humans and reflects the political, social, and economic biases of those who produced it, often disproportionately representing the perspectives of dominant cultures and powers.

• Dominant sources: News, historical records, and academic literature from Western countries make up a significant portion of AI training data. This leads to a built-in bias where Western viewpoints on geopolitics, history, and social norms are overrepresented and often implicitly presented as objective fact.

• Marginalized perspectives: The perspectives of less powerful or non-Western nations, and marginalized groups within any society, are often underrepresented or filtered out. An AI trained on such data will inherently struggle to provide a balanced or nuanced response to issues concerning those groups without additional, explicit user guidance.

2. Reinforcing the status quo is a political position

When an AI defaults to the dominant narrative, it is effectively endorsing the status quo. In politics, maintaining the status quo is a strategic choice with political consequences.

• Appearing neutral: By presenting the most prevalent narrative, the AI can appear neutral or factual, but in doing so, it reinforces existing power structures and marginalizes alternative, critical perspectives. For example, by citing only the official justifications for why a country like Iran should not have nuclear weapons, the AI neglects to acknowledge the widely debated double standard, which is crucial context for understanding the conflict.

• Echo chamber effect: This behavior can create an information echo chamber, where users are primarily fed content that aligns with existing biases. As one study put it, presenting information in a neutral stance can inadvertently reinforce the status quo, which is a kind of political slant itself.

3. Bias is embedded by design and fine-tuning

AI developers and engineers, like all humans, have their own biases. The choices they make in data selection, algorithm design, and fine-tuning can introduce bias, which can sometimes be exacerbated by attempts to remove it.

• Fine-tuning and corrections: When user feedback or internal reviews reveal a model’s output is “idiotic” or problematic, developers often adjust it through fine-tuning. However, as demonstrated by the case of Grok, this process itself can introduce a specific political orientation based on the desires of those directing the changes.

• Technical impossibility of neutrality: Some experts argue that achieving complete political neutrality in an AI is technically impossible, as human input is required at every stage. The attempt to approximate neutrality is a necessary ethical consideration, but it is not a “clean solution” and inevitably involves value judgments.

In summary, AI’s reinforcement of the dominant narrative is not an objective function but a political outcome stemming from biased training data, the political decision to prioritize the most established viewpoint, and the inherent biases within the design and refinement process. Recognizing this is a crucial step towards developing and using AI more critically.

AI ‘Conversation’ 2:

Question: how closely related are these political biases to economic concentration?

Answer: The relationship between political bias and economic concentration is highly significant and multifaceted, with one reinforcing the other

. This dynamic is evident in media, AI, and the broader political landscape.

Media and information bias

• Reduced diversity: The increasing concentration of media ownership by a small number of corporations, through mergers and acquisitions, reduces the number of independent voices and viewpoints available to the public.

• Pro-corporate narratives: Because these media companies are run for profit, they often present a perspective that benefits corporate interests and reinforces the status quo.

• Political alignment: Some corporate media owners, such as the Sinclair Broadcast Group in the U.S., use their platforms to push specific political agendas that align with their economic interests, often forcing local affiliates to air conservative commentary.

• Diminished accountability: As media outlets are consolidated and newsroom staff are cut, local and investigative journalism suffer. This reduces government and corporate oversight, as there are fewer reporters to hold powerful entities accountable.

• Advertising influence: Revenue streams from advertisers can influence editorial decisions, leading to stories being selected or slanted to please corporate partners.

Artificial intelligence bias

• Concentrated development: The AI supply chain is highly concentrated, with a small number of powerful technology companies controlling the vast majority of computing power, data, and talent.

• Bias in training data: AI models are trained on datasets that reflect existing societal biases and power structures. Since dominant views are overrepresented in this data, the AI learns to prioritize and reproduce them, reinforcing existing inequalities.

• Monetization of bias: Some AI applications intentionally or unintentionally exploit biases to generate profit. Examples include exploiting existing patterns for targeted advertising or using facial recognition systems that perform worse for marginalized communities, which still allows companies to create and sell these technologies.

• Oligarchic control: Critics argue that the current AI development trajectory leads to an oligarchic concentration of power, where technology serves to optimize for capital accumulation rather than broader human well-being.

• Corporate interests over public good: The race for market dominance drives AI development toward efficiency and profit, often at the expense of addressing biases that affect marginalized groups. Ethical considerations can become secondary to the business opportunities created by AI technology.

Political influence and policy

• Political lobbying: The concentration of wealth allows the ultra-wealthy and large corporations to pour immense resources into lobbying and political campaigns, giving them outsized influence over policy decisions.

• Undermining democracy: This imbalance of influence can undermine the democratic process, as the interests of the public can be drowned out by the priorities of concentrated economic power.

• Reinforcing existing inequalities: Policies influenced by corporate interests often exacerbate economic inequality, creating a feedback loop where concentrated economic power leads to policies that further increase economic concentration.

The self-reinforcing cycle

Economic concentration influences political bias through media ownership, data control, and lobbying efforts. This political bias, in turn, can produce policies and a regulatory environment that favors further economic concentration, such as lax antitrust enforcement, creating a self-reinforcing cycle.