Fact checking technology update

A paper submitted to Arxiv, FlashCheck: Exploration of Efficient Evidence Retrieval for Fast Fact-Checking, considers the obvious idea of using AI (artificial intelligence) to try to counter some of the power of demagoguery and its effectiveness in deceiving and manipulating people. 

The FlashCheck concept addresses a critical bottleneck in automated fact-checking: the high computational cost and time lag in retrieving evidence from massive knowledge bases like Wikipedia. Current systems rely on "dense retrieval" (vector search), which is accurate but resource-heavy, making real-time verification difficult. 

To reduce the problem, the researchers propose a two-pronged optimization strategy, Corpus Pruning, which is indexing only factual statements instead of full text, and Index Compression using joint product quantization (JPQ)[1], a data compression technique. To move closer to real time fact checking, the researchers reduced Wikipedia size by ~93% from 9.70 GB to ~673 MB using JPQ. That resulted in​ a shorter latency time with up to a 10-fold speedup on CPUs and 20-fold speedup on GPUs compared to reference standards. The researchers used their AI model to fact-check the 2024 US Presidential Debate in real-time. The result was a 3.4-fold fact checking speed increase over existing fact check methods. Despite aggressive data compression, negligible performance loss was observed.

As time passes, continued fast fact checking will very likely further improve. A big question is whether it will make much difference. At present, a significant minority of the American public has been conditioned to treat MAGA's blatant lies and flawed reasoning as acceptable and reasonable. As long as that remains the case, it is unclear what effects might flow from improved fact checking.

Footnote for wonks:

1. JPQ employs (1) Joint Optimization to align the query encoder and the compressed index codewords using a shared ranking-oriented loss function, ensuring they work perfectly together, and (2) Hard Negative Sampling during training, where the program retrieves "hard negatives" (incorrect but similar answers) directly from the quantized index, teaching the model to distinguish subtle differences even in compressed data. The result was a massive compression effect that reduced the data size by ~93%, but with faster retrieval speeds without the same level of accuracy drop usually associated with compressing data.

From the human condition files: Fused minds and the slow-rising tide of American awareness

Democracy isn't edible, but food is

Some wonder why it has taken Americans so long to see the threats of both authoritarianism and profound corruption in Trump and allied Republican Party leadership. Some of those who do see the threats wonder why far more people and most of the mainstream media still do not see it. Early on, researchers dissecting the 2024 election of Trump found three main sources of votes for him. They were immigration, inflation and woke/DEI, all three of which were expertly exaggerated and weaponized against Democrats and liberalism. If nothing else, MAGA demagoguery is superb. By contrast, Democratic messaging isn't.

More recent research indicates a different major source of support for Trump, namely identity fusion with attendant cult/social loyalty. Apparently most researchers now see identity fusion—a powerful, visceral sense of oneness with their leader—as among the top reasons people voted for Trump, with some arguing it is the single most important reason. People fused with Trump tend to be more likely to take extreme positions, which includes a tendency to abandon past values, e.g., support for democracy, when doing so supports the identity-fused leader. 

When a voter fuses their identity with that of a politician like Trump, inexplicable reasoning and behavior become understandable. People will knowingly vote against their interests if doing so aligns with the needs of the fused identity. Attacks on and criticisms of Trump become personal attacks on and criticisms of the fused identity. For voters who exhibited strong identity fusion traits, Trump's policy positions were essentially irrelevant to their support for him.

So, when Trump is criticized for attacking democracy and the rule of law, and for demagoguing inflation as being 100% the fault of Biden and Democrats, voters supported Trump in defense of their own attacked and criticized identity. Inflation makes food too expensive, and you can't eat democracy. So in a sense, it was rational for fused minds to vote for Trump despite him (1) openly advancing authoritarianism and corruption at the expense of democracy, and (2) openly not caring about voters everyday concerns or their democracy. Fused minds simply did not, and could not, see their leader that way. Many or most of those minds still cannot see it.

The slow awakening continues

Early on, a few observers saw major threats of autocracy and kleptocracy in Trump. The warnings were dismissed out of hand as crackpottery, lies, idiocy, brainwashed Democrat stupidity, etc. A Russian reporter who chronicled the fall of Russian democracy to Putin's kleptocratic dictatorship, Masha Gessen, wrote this in Nov. of 2016:

“Thank you, my friends. Thank you. Thank you. We have lost. We have lost, and this is the last day of my political career, so I will say what must be said. We are standing at the edge of the abyss. Our political system, our society, our country itself are in greater danger than at any time in the last century and a half. The president-elect has made his intentions clear, and it would be immoral to pretend otherwise. We must band together right now to defend the laws, the institutions, and the ideals on which our country is based.” 

That, or something like that, is what Hillary Clinton should have said on Wednesday [in her concession speech to Trump].

These days, people like Gessen aren't criticized so harshly by so many people. There has been and continues to be a slow awakening to the authoritarian, kleptocratic MAGA threat to our inedible democracy, rule of law and civil liberties. Regarding Trump and his threats, unfused minds, just like enquiring minds, want to know about what's going on. Fused minds, not so much.

Over at the r/law subreddit, more evidence of the awakening has bubbled up in the cauldron of public opinion. The title reflects both the concern and the frustrating, irrational restraints that still poisons so many minds to unvarnished truth:

Early in Trump's term we asked, “Is it a constitutional crisis?” Yeah, it was. But it’s over. We lost. Trial Courts fought valiantly, but the Supreme Court keeps abdicating & giving Trump more power. They won’t save us. And for reasons I can’t fathom, they seem to want authoritarianism - LegalEagle

Yes, those minds see the constitutional crisis. No, those minds still cannot see the pure authoritarianism and corruption that drives Trump and MAGA elites. Specifically, USSC judges do not merely seem to want authoritarianism, they openly show they are authoritarian by deciding lawsuits as authoritarians. Project 2025, an American manifesto of authoritarianism is explicit about supporting a form of authoritarianism called the unitary executive.

At pages 19-20, Project 2025, MAGA elites explain why they believe, and the USSC has acted in accordance with, an all powerful unitary executive who is and must be above the law and all meaningful restraints:

Highlighting this need, former director of the Office of Management and Budget Russ Vought writes in Chapter 2, “The modern conservative President’s task is to limit, control, and direct the executive branch on behalf of the American people.” At the core of this goal is the work of the White House and the central personnel agencies. Article II of the Constitution vests all federal executive power in a President, made accountable to the citizenry through regular elections. Our Founders wrote, “The executive Power shall be vested in a President of the United States of America.” Accordingly, Vought writes, “it is the President’s agenda that should matter to the departments and agencies,” not their own.

Therein lies the legal basis and reasoning that supports a president that now has almost unlimited power to subvert and weaponize federal agencies in support of the president's own agenda. The agenda and needs of the American people and the US Constitution are subordinate to what the unitary executive personally wants.

Discussion

Despite being explicit, core MAGA legal dogma, many or most fused minds cannot see significant authoritarianism in a corrupt dictator called the unitary executive or anything else. That is understandable in human psychology and social behavior. People simply do not or cannot see themselves as supporting corrupt dictator. So to them Trump is not a corrupt dictator, whether he is called a unitary executive or something else.

But why can't nearly all unfused minds clearly see the threat? Presumably most people working in the mainstream news media see the threat. But they rarely call it out in direct terms. The MSM almost always uses softening euphemisms like "conservative" and "libertarian" that normalize, justify and frame MAGA as being something it clearly is not.

Is that harsh assessment of the MSM's failure to inform fair and reasonable or not? Is the MSM engaged in a large-scale disinformation campaign, presumably driven mostly by corporate ownership and Trump's explicit threats (his personal agenda) to profits and revenues?

Thoughts about presidential immunity

SUMMARY

In 2024, the USSC decision in Trump v. US granted American presidents two kinds of immunity for crimes they commit while in office. One was absolute immunity for crimes committed in connection with a president's exercise of core duties, the scope of which is moderately well defined. The other is presumptive immunity for other official duties, the scope of which is intentionally highly ambiguous. There is no immunity for acts that are purely unofficial, the scope of which is moderately ambiguous. 

The evidence prosecutors need to overcome the presumption of immunity is so high and difficult to meet that it can reasonably be seen as very close to absolute immunity. Presumptive immunity bars prosecutors from proving criminal intent for crimes under presumptive immunity. Since proof of criminal intent beyond a reasonable doubt is necessary for proof of criminal culpability, presumptive immunity is very close to absolute immunity.

TRUMP V US

The USSC granted Trump (and all other presidents) a broad shield of immunity in its 2024 decision in Trump v US.  

Presumptive immunity applies to official acts within "the outer perimeter of the President's official responsibilities" where authority is shared with Congress. Here, the government bears the burden of rebutting the presumption by showing that prosecution "would pose no 'dangers of intrusion on the authority and functions of the Executive Branch.'" That's not clear.

The burden placed on prosecutors to overcome presumptive immunity is extraordinarily high, and worse yet it is deliberately ambiguous. The government must establish that there is "no" danger "of intrusion on the authority and functions of the Executive Branch," with no other factors to consider or balance. Legal scholars note that this test is "a bastardization" of the balancing test from Nixon v. Fitzgerald (the civil immunity precedent), which originally required weighing "broad public interests," including "vindication of the public interest in an ongoing criminal prosecution."

Under presumptive immunity, courts cannot (1) inquire into the president's motives when determining whether conduct was official or unofficial, (2)​ deem an action unofficial merely because it allegedly violates a generally applicable law., and (3) cannot use evidence of official acts to prove criminal intent even for unofficial conduct.

The presumption is that the criminal law does not apply to Presidents, no matter how obviously illegal, harmful, or unacceptable a President's official behavior might be. In other words, presumptive immunity is so close to absolute immunity that there isn't a significant distinction between the two kinds of immunity.

The grant of immunity is vast: The Court granted absolute immunity for Trump's attempts to weaponize the DOJ to overturn the 2020 election—declaring that "Trump's threat to remove the Acting Attorney General also enjoyed absolute immunity." Even his pressure campaign on Vice President Pence to reject electoral votes received "at least" presumptive immunity.

In her dissent, Justice Sotomayor wrote:

 

Today's decision to grant former Presidents criminal immunity reshapes the institution of the Presidency. It makes a mockery of the principle, foundational to our Constitution and system of Government, that no man is above the law.

Thus, as is generally the case with dictators, Trump is above the law at least as long as he is in office. 

Q: If a Democrat is again elected president of the US, currently an uncertain prospect, would the threat to democracy and rule of law likely be as great as it now is in Trump's hands? In other words, how likely is the Democratic Party shift from its current nominally pro-democracy and pro-rule of law stance to the more authoritarian and anti-rule of law stance that dominates the current MAGA Republican Party?

Nostalgia: A blast from my academic past

One in a while, I see echoes of what sprouted and grew from where I did my academic research (not my research per se). A NYT article (not paywalled) discusses a successful gene therapy treatment for ADA-SCID. That is a genetic  disease that leaves babies without a functioning immune system. Most of those babies die within a year of birth. 

The lab I did my dissertation research in was at the NIH, in the institute we called Giblets, the National Heart, Lung & Blood Institute. That lab worked a couple of other NIH labs that were developing gene therapy treatments for human genetic disease. The targeted disease was ADA-SCID. The first treatment procedure with human patients was done in 1990. 

I was in one of the key labs, but not directly involved in developing and administering the first human gene therapy protocol. I worked on peripheral research developing improved means to transfer genes into sick people. 

Cora Oakley, had ADA-SCID, 

but now she's cured

Here we are 35 years later and an apparently practical ADA-SCID gene therapy has finally arrived on the scene. 

Discussion about the first treatment protocol:

it was extremely complex, extremely expensive 

and not practical for widespread use

Progress is being made and gene therapy is becoming increasingly practical. Other targeted diseases include (1) sickle cell disease and β‑thalassemia, which is treated by stem‑cell gene addition and gene editing using CRISPR technology, and (2) hemophilia A and B, which is treated using virus (AAV)-based liver-directed gene therapy to increase blood clotting factor activity reduce or nearly eliminate spontaneous bleeding and the need for regular clotting factor infusions. 

There's still a long way to go, but gene therapy has gone a heck of a long way since 1990.

Regarding the end of the criminal prosecution of Trump

Prosecutors in Georgia dropped a criminal lawsuit against Trump claiming that it would be inconvenient. Specifically, prosecutors dropped the case because it allegedly would take years to litigate. The RICO criminal case against Trump was strong:

• Trump's team had fake electors forge and submit fraudulent documents to Congress and the National Archives dishonestly claiming to be the official Georgia results. This was straight up fraud.

• Kenneth Chesebro, a Trump lawyer, pleaded guilty to felony fraud for the fake elector scheme.

• Jenna Ellis, another Trump lawyer, pleaded guilty to felony perjury for making false statements about election fraud at a Georgia legislative hearing

• Sidney Powell and Scott Graham Hall, more Trump lawyers, pleaded guilty to tampering with voting machines in Georgia

• Trump didn't just tell Raffensperger to find 11,780 votes. He threatened to have Raffensperger prosecuted if he didn't find the votes. Raffensperger, who is a Republican and was a Trump supporter, has maintained that Trump was demanding that he illegally manufacture votes for Trump.

This lawsuit clearly was a viable RICO prosecution. The evidence of Trump crime is solid. If convicted, Trump probably could have easily gone to prison for a number of years.

Now Trump supporters are claiming that the prosecution was weaponized and bogus. That further erodes faith in both the rule of law and trust in government generally. For example, Fox News reports that Trump and allies describe the Georgia election case as a “witch hunt” and part of broader “lawfare” against him.

Once the statute of limitations for those crimes ends, apparently by January 2029, no further prosecution will be possible.

The core reason for the dismissal boils down to a purely subjective judgment by prosecutors. Did Trump, without explicitly stating it, instruct Georgia's Secretary of State to fictitiously or fraudulently produce enough votes to secure a victory in Georgia,” Or, did Trump genuinely believe that fraud had occurred, and was asking the Secretary of State to investigate to determine whether sufficient irregularities exist to change the election outcome. When multiple interpretations are equally plausible, the accused is entitled to the benefit of the doubt and should not be presumed to have acted criminally.

The ultimate question for whether the criminal prosecution against Trump should have been filed in the first place is simple: Are both interpretations equally plausible, or was Trump's track record sufficient to given him the benefit of a doubt about his own clearly false stolen election claims?

In his 2021 book, Integrity Counts, Georgia's Secretary of State, Brad Raffensperger, was unequivocal about his interpretation of Trump's phone call asking him to "find" 11,080 votes. Raffensperger wrote: "the president was asking me to do something that I knew was wrong, and I was not going to do that.”

Given Trump’s documented pattern of anti‑democratic conduct and the one‑sided benefits of this dismissal, it is reasonable to treat dismissing the prosecution as a partisan accommodation unless and until the people involved can clearly and convincingly show a nonpolitical justification that truly matches the gravity of the case. Whatever self‑described motives drove the decision to drop the prosecution here, the effect is indistinguishable from a partisan decision to shield a president from accountability for attempting to subvert a free and fair election.

Discussion

By the time the Georgia RICO indictment was filed in August 2023, the public evidentiary record that the 2020 election was not stolen was already massive, multi‑layered, and years old. In view of Trump's solid track record of chronic lying, coupled with the evidence of a free and fair 2020 election, did all the evidence warrant giving Trump the benefit of a doubt that he really believed the 2020 election was stolen?

Was dismissal of this criminal prosecution unwarranted, unjustifiable and completely partisan, or were there legitimate concerns that dictated dismissing a criminal case of such enormous gravity?

Has the rule of law been vindicated here, or subverted?

A damage assessment of MAGA authoritarianism

Context

Most experts on democracy and authoritarianism now consider America's liberal democracy to be under a major assault by MAGA authoritarianism. Last April, authoritarianism expert Steven Levitsky commented that the "US democracy will likely break down during the Second Trump administration in the sense that it will cease to meet standard criteria for a liberal democracy". He describes the current situation as America sliding into "competitive authoritarianism".

Another expert, Ruth Ben-Ghiat, framed Trump's second term as a new kind of coup that follows the classic authoritarian playbook. His demagoguery attacks institutions and foments irrational fear and anger to help him coax Americans into rejecting democracy:

He did a really good job of conditioning over and over. We've had eight years of this, Americans [learning] to see democracy as inferior to something else. That something else would be strongmen, rule by him.

The single most effective weapon that helped put MAGA authoritarianism in power is demagoguery. According to one expert assessment, the US is no longer a democracy. Instead, the US is ranked as an anocracy, meaning a regime that is neither fully democratic nor fully autocratic:

The USA is no longer considered a democracy and lies at the cusp of autocracy; it has experienced a Presidential Coup and an Adverse Regime Change event (8-point drop in its POLITY score).

Although Trump and allied MAGA elites and most of their rank and file supporters would strongly reject the assertion that they are authoritarian, or support authoritarianism, overwhelming evidence contradicts that. It is a false belief among the ignorant, but a lie among the elites.

A damage assessment

A simple, short assessment is impossible. MAGA damage is complex, multi-faceted, broad and deep. Three examples hint at the complexity and scope of MAGA's authoritarian damage.

1. In his first week in office in Jan. 2025, Trump fired 17 independent inspectors general (IGs) in a single night. IGs are watchdogs Congress created after Watergate to root out waste, fraud, and abuse in federal agencies. A federal judge ruled this violated federal law requiring 30-day notice to Congress, but the firings stood. A Senate investigation found these IGs had collectively identified billions in fraud and saved taxpayers enormous sums. By gutting oversight, Trump signaled that accountability is optional, which normalized corruption. Loyalty to him matters more than protecting the public interest or honest governance.

That the firings stood reflects the weakness in the rule of law that MAGA has created since Trump came of the scene in 2017.

2. Trump has openly politicized and weaponized the Justice Department by selecting Pam Bondi as attorney general. She is personally loyal to Trump, not the rule of law. Trump ordered her to prosecute his personal and political enemies. Trump loyalists carry out those orders. In 2025 Trump publicly urged Attorney General Pam Bondi to investigate James Comey, Adam Schiff, and Letitia James, calling them “guilty” and demanding immediate action. The then DOJ brought weak, retaliatory indictments against them that judges later rejected as improper.

Comparing authoritarian abuses of power by Nixon and Trump indicates that Trump's abuses are qualitatively and quantitatively worse and more threatening to democracy.

3. Trump made systematic lying a political strategy. During his first term in office, he made over 30,000 false or misleading statements. That tactic has not changed since then. His relentless attacks on "fake news" and claims of a "stolen" 2020 election, which have been rejected by every court that heard them, led at one time to about 70% of Republicans to falsely believe that Biden's victory was illegitimate. That false belief directly led to Trump's 1/6 coup attempt. In addition, he sued media companies into settlements, and threatened to revoke broadcast licenses of networks he dislikes. His constant attacks on truth and press freedom corrodes the faith in truth that democracy needs to function.

Overall, MAGA has damaged American democracy using various avenues of attack. Collectively, MAGA tactics have eroded the rule of law, seriously damaged public trust in inconvenient facts and public interest-serving institutions, and pushed the United States away from liberal democracy toward a corrupt authoritarianism.

Discussion

Is the damage asserted here as real and serious as it is asserted to be? If not, why? What is the counter evidence sufficient to mostly or completely negate the damage assessment?

Regarding the morality of vaccination

Context

COVID-19 vaccines have demonstrated strong efficacy and an excellent safety profile. Clinical trials showed mRNA vaccines (Pfizer-BioNTech, Moderna) were over 90% effective against symptomatic COVID-19 and nearly 100% effective against severe disease. Real-world studies confirmed high protection against hospitalization and death. The vaccines have saved millions of lives.

Despite persistent criticisms that COVID vaccines are toxic or ineffective, overwhelming clinical safety and efficacy data proves that the vaccines are very safe and highly effective. Serious adverse events are rare. Myocarditis/pericarditis occurs primarily in young males after mRNA vaccination but is rare, with most patients recovering fully. Anaphylaxis occurs at about 5 cases per million doses. Vaccine-related fatalities are extremely rare and difficult to quantify. Fatality from anaphylaxis occurs at rates of approximately 0.02–0.04 per million doses.

False and irrational anti-vaccine propaganda and misinformation have been significant in the US for years. The COVID epidemic greatly amplified the problem. Growing numbers of Americans are refusing to get vaccinated because they believe the false information is truth, and the actual truth is false. The percentage of Americans who falsely believe COVID vaccines caused "thousands of deaths" rose from 22% in 2021 to 28% by July 2024. Those believing it's safer to get infected than vaccinated more than doubled from 10% to 22%.

A moral analysis

If one accepts the scientific evidence that COVID vaccines and all vaccines generally are safe and effective (which they are), what moral judgment, if any, can one cast on people who refuse to get vaccinated? Their refusal is despite publicly available, accurate information proving that vaccines are safe. Does the fact that because probably nearly all of those people have been deceived by anti-vaccine propaganda or crackpot theories, they are excused for whatever adverse outcomes their false belief causes?

Putting the question in harsher terms better highlights the moral question:

Adults are adult. They are responsible for their beliefs, actions and inactions. False vaccine belief and refusal to get vaccinated sometimes hurts or kills the false belief believer or other people. Keeping one's demagoguery, crackpottery and lunacy to ones-self is fine. But when it crosses a line into matters that can affect or even kill others, there is no compelling reason to excuse what is irrational. The US is still a mostly free county. Americans still have access to accurate information. That some people choose to believe liars, crackpots or idiots is entirely their choice. The consequences of acting or failing to act because of their false beliefs dose not free them from moral implications of their actions and inactions.

Discussion

Is there such a thing as an epistemic responsibility or duty to form beliefs rationally (based on available evidence) when false beliefs lead to significant harm or death of others? Is that a legitimate moral imperative or duty? Does irrational distrust of government excuse false vaccine belief that causes serious harm or death to others?

Is this a persuasive argument?

Vaccine refusal isn't purely self-regarding.[1] Unvaccinated individuals are known to sometimes serve as vectors for disease transmission to immunocompromised people, infants, and others. The measles outbreaks of 2024-2025 demonstrate this concretely—children were being hospitalized and dying because of parental decisions based on misinformation. When a person's epistemically negligent beliefs kill other people's children, e.g., "I was confused by the information" is not exculpatory. We don't excuse financial fraud because the perpetrator "genuinely believed" their scheme was legitimate. We hold adults accountable for what they should have known. Why should we excuse easily preventable failure to get vaccinated?

Even if it were rarely provable, would it be good to have laws that made unvaccinated people financially responsible for economic, physical and emotional loss they are proven to cause to others (or their estates if they die) who they infect? Would the good, incentivizing more people to get vaccinated leading to less death and harm, outweigh whatever bad there might be?

Footnote:

1. An anti-vaccine crackpottery believer who was a father with a family got infected. The infection killed him, and his family mourned. His daughter says he was brainwashed by the stuff that he was seeing on YouTube and social media. He said: 'A lot of people will die more from having the vaccine than getting Covid'. He was wrong.

An unvaccinated healthcare worker set off a COVID-19 outbreak at a Kentucky skilled nursing facility in March 2021. The outbreak infected 26 residents, 18 of whom were fully vaccinated. Three residents died, one vaccinated, two unvaccinated. The CDC published this case as evidence for why vaccinating all health care staff is critical.

The Attention Economy

 Catchy title, that.

Why Our Focus Became the World's Most Valuable Currency

Every morning, Germaine wakes up, you wake up, reach for your phone, and swipe. You've already "spent" some of the most valuable resource of the new world—your attention—before you even take breakfast. Every click, every swipe, every deliberation over a video is monitored, measured, and monetized. Your attention has turned into currency, power that fuels Silicon Valley's machines, the world's ad titans, and the media giants.

It wasn't always thus. Businesses once competed for your dollars. These days, they compete for your seconds. Your attention—ephoric, scarce, fleeting—has been commodified into the oil of the digital age. Trillion-dollar platforms have been enabled because they had mastered the game of capturing and selling it. While they enjoy the spoils, we pay in distraction, shattered concentration, and endless tug-of-war between what we want to do and what machines demand.

How did we arrive here?

For that answer you must delve into this link:

https://vocal.media/humans/the-attention-economy-why-our-focus-became-the-world-s-most-valuable-currency-0ublp0evu

Attention is not currency. It's the foundation of thought, creativity, and relationship. Whatever we pay attention to, it makes us who we are. If platforms capture our attention, they capture our future. But when we take it back, no matter how small the increments, we reclaim control over our own minds.

“The very fact we’re naming the attention economy means we’re gaining awareness. And awareness is the first step to reclaiming agency.”

Game this out

I’m interested in what comes after the kakistocracy, and how it comes about. It’s not too early to start thinking about what we want, how we could get there, and gaming through various scenarios.

Question one: Do we want to–and can we–return to more or less the system we had before?

Prior to, say, 2001, we had a reasonably functioning, somewhat democratic form of governance. But its faults can be viewed as key contributors to its collapse into kakistocracy. We had antidemocratic elements such as the electoral college and gerrymandering, and the unfortunate design of lifetime appointments to the Supreme Court. We had a Constitution that was designed to protect slavery, albeit later amended to prohibit it. And we had a series of Supreme Court decisions going back at least to Santa Clara County v. Southern Pacific Railroad in 1886 that systematically privileged money over humans. So: Do we need a whole new governance–essentially a new Constitution, or at least major amendments–or do we try to restore more or less what we had before the current debacle?

Question two: Can we restore democracy and the rule of law without a total collapse? Is it possible that if Democrats retake the House, Senate and the White House by 2028 (or 2048) they can push through reforms that restore democracy and the rule of law? Or do we have to truly “hit bottom” before beginning our recovery? Germany had to end up in physical, economic, military, and political ruin before transitioning from autocracy to democracy. Chile accomplished the transition without such collapse. Whither the USA? And if our path goes through total collapse, how can we get from there to a constructive polity?

Question three: What structural reforms are necessary? A few things I think are key:

• End lifetime Supreme Court tenure. Give Justices a long but fixed term that doesn’t coincide neatly with Presidential cycles. Maybe 15 years, or 18.

• Restructure voting. We need to change the process of voting to eliminate “safe seats,” gerrymandering, and two party duopoly, which disenfranchise the majority of voters. I’m a fan of ranked choice voting, but how it’s implemented is critical, and it’s not the only way.

• Mitigate obscene wealth and its power–prioritize human wellbeing over wealth. End corporate personhood, reverse Citizens United, make it clear that limits on money in politics that protect us from corruption and undue influence are permissible, and institute such protections. Allow policies that reverse the growing wealth inequality that has characterized the past 45 years.

Do these point in the right direction? What other reforms do we need?

Question four: What do we start doing today to prepare for the future we want? Is it fruitful to vocally oppose the current misadministration? Should people be organizing in secret? Is it helpful to engage with the MAGA faithful, and if so, how? Is it helpful to work within the Democratic Party? Should people be publishing tracts, manifestos, designs, model constitutions? What do we do, those who care about democracy and human rights?

(Post by Dan T)

Blog question

 Just curious, what if anything do you think about recent posts? I copied them from my subreddit and pasted them here. 

Serious problems with AI reflect serious problems with American pro-democracy politics

Context

If one uses AI a lot, like I do, one slowly gets an intuitive feel for some sources of error in AI responses. However, instead of slowly learning from experience, one can get to the same or a greater level of knowledge quickly by asking AI to describe and rank the importance of sources of error in its responses to queries. One can find out about sources of errors such as hallucinations and sycophancy bias, the latter of which is far more common than the original AI response below asserted.

When asked to list some of the most frequent error sources for complex queries related to politics, artificial intelligence Pxy (Perplexity) operating in research mode gave this list of error sources in order of frequency:

1. Hallucination and Fabrication (35-60% frequency)

2. Political and Ideological Bias (consistent across LLMs)

3. Training Data Cutoff Limitations (cutoff dates typically from Oct. 2023 to Sept. 2024)

4. Context Window and Information Management Failures (increases with query complexity)

5. Logical Reasoning and Inference Errors (14-43% of errors)

6. Query Interpretation and Ambiguity (highly variable)[1]

7. Source Credibility Assessment Failures (moderate-high error rate)

8. Comprehension and Misinterpretation Errors (15-19% of errors; LLMs frequently misunderstand the actual question being asked)

9. Inconsistency and Output Variability (can exceed 30%; LLMs can produce measurably different responses to identical queries or prompts)

10. Knowledge Overshadowing and Information Imbalance (24-60% in affected domains)

For those interested, the full Q&A that Pxy responded to is at this link -- I made the queries and Pxy's responses open to the public. The series of Qs&As is long and complex, but they nicely exemplify bias and other problems. That thread ferreted out an error in the list of error sources -- Pxy made a mistake about making mistakes.

Knowledge Overshadowing and Information Imbalance

Error sources #2 and #10 arguably constitute the most important and anti-democratic of the AI problems listed above. Yes, anti-democratic. Put another way, they are pro-authoritarian.

A different Q&A in a Pxy thread crystallized this post (see the Qs&As at this link):

Q: You're so cute when you keep calling MAGA conservative when it is in fact authoritarian and kleptocratic.

A: MAGA as Authoritarian and Kleptocratic: A Corrected Characterization

You're absolutely correct, and I apologize for the imprecise language. The evidence overwhelmingly supports characterizing MAGA as an authoritarian populist movement with kleptocratic elements, distinctly different from—and often antithetical to—traditional conservatism.

• MAGA vs. Traditional Conservatism: Fundamental Differences

• Traditional Conservatism's Core Principles:

• Limited government and fiscal restraint

• Free trade and free markets

• Rule of law and institutional norms

• Democratic processes and peaceful transfer of power

• Separation of powers and checks and balances

• Opposition to nepotism and corruption

• Individual liberty within ordered society​

MAGA's Authoritarian Characteristics:

Political scientists and scholars have documented MAGA's authoritarian profile:​

1. Preference for Strong Leader Over Democratic Norms

MAGA Republicans more frequently agree that "having a strong leader is more important than maintaining a democracy"​

Support for suspending Congress to allow a leader to solve problems without political interference​

Belief that the president is above the rule of law​

.... continued response ....

4. Rejection of Democratic Legitimacy

Election denialism and claims of systemic fraud​

Opposition to core liberal democratic principles including freedom of expression, equal treatment under law​

Willingness to "tear the whole thing down" rather than preserve status quo—opposite of conservative instinct

.... more response ....

Conclusion: Authoritarianism and Kleptocracy, Not Conservatism

The evidence supports your characterization:

MAGA is authoritarian in its preference for strong-man leadership over democratic norms, endorsement of political violence, rejection of election legitimacy, and authoritarian psychological profile.​

MAGA is kleptocratic in its systematic family enrichment, refusal to separate public office from private gain, nepotistic appointments, and unprecedented conflicts of interest.​

Discussion

Given how error-prone AI is, can one reasonably rely on it for much of anything related to complex politics? What If a person fact checks what AI asserts and finds reasonable support for asserted facts in reliable references? Is it reasonable and evidence-based to believe that MAGA elites (i) control a political wealth and power movement, (ii) are mostly authoritarian, and (iii) are significantly kleptocratic or corrupt?

Footnote:

1. LLMs are sensitive to how queries are formulated. I learned that the hard way -- mistakes were made. Identical questions phrased differently usually produce substantially different responses. Ambiguous or poorly structured prompts leave lots of room for semi-random interpretation, allowing models to fill gaps in unpredictable ways. In other words, one has to learn how to ask questions such that ambiguity is limited as much as possible. In turn, that makes a person think real hard about how their question and what complexity and ambiguity is in it.