Trump accuses Beijing of the largest electoral data theft in history
Unemployed engineers, hired philosophers: the moral restructuring of artificial intelligence
Brief introduction
The phrase that states have no permanent friends or enemies, only permanent interests, has been repeated a thousand times. Many apply this to geopolitics, but the truth is that the game is not always played on maps, but within consciences. In my view, consciences should play the most significant role in geopolitics, but I suppose I can only afford to say that because I merely write about geopolitics rather than being a protagonist within it. Today is one of those days. Two seemingly incomparable pieces of news share the exact same root, and that root is the deepest of all: who decides what a society holds to be true, and who decides what a machine will hold to be good.
The first came from Washington in the early European hours. President Trump accused the People’s Republic of China of executing the largest theft of electoral data in history and declassified, during prime time, a set of intelligence documents to prove it. The second has not made any headlines, yet it deserves to make them all: while newly qualified software engineering graduates face a near 7.5% unemployment rate in the United States, major artificial intelligence laboratories are competing for philosophers with six-figure salaries. The first story belongs to the old art of political warfare. The second decides what morality will govern systems that will very soon select military targets. Both aim at the same bullseye: human discernment.
We devote today’s entire report to these two matters. And we warn from the very first line that neither of them tolerates the aseptic treatment so beloved by the mediocre political class of our century—the type that mistakes prudence for abstention, or the paralysis of a deer in headlights for balance, or neutrality for cowardly silence.
Trump accuses Beijing of the largest electoral data theft in history
The Facts
President Donald Trump addressed the nation from the East Room of the White House on Thursday, 16 July at nine o’clock in the evening, Eastern Time (early Friday morning in Europe). He announced the immediate declassification of a substantial volume of intelligence material, asserting that over several years, beginning with the 2020 election cycle, the People’s Republic of China carried out what is believed to be the largest compromise of electoral data in history, illicitly acquiring 220 million American voter files. The files allegedly contain names, addresses, telephone numbers, voting histories, political affiliations, and military status. The timeframe targeted by the White House spans from 2020 to 2023; some accounts extend into 2024. These documents have been published on the official White House website.
Trump added that federal agencies first became aware of the voter registry compromise in 2020, when they discovered that data belonging to tens of millions of voters across eighteen states had been purchased, stolen, or hacked by China. He then levelled a second, internal accusation: that members of the intelligence community actively suppressed this information, hiding it from both the president and the American public, even removing it from the President’s Daily Brief. He further maintained that an analysis by the Department of Homeland Security (DHS) identified 278,000 non-citizens registered on federal voter rolls, a figure he speculatively raised to one million. He closed by demanding that Congress pass the Save America Act, which would mandate photo identification and proof of citizenship to vote.
Crucially—and this is a point no serious analyst can omit, as it forms the crux of the entire matter—the declassified materials contain no accusation whatsoever that votes were altered, that voting machines were hacked, or that any election result was changed due to foreign interference. Neither in 2020 nor in any subsequent cycle. The accusation is one of mass data harvesting, not the manipulation of election counts. These are two radically different concepts that must never be conflated.
Image of the President of the United States, Donald Trump, participating in events at the Great Hall of the People and greeting the President of the People’s Republic of China, Xi Jinping, on 14 May 2026 in Beijing (China), during a visit focused on trade, regional security and strengthening bilateral relations between the two largest economies in the world – PHOTO/ Kenny Holston via REUTERS
The counterweights are equally verifiable and must be laid out. In numerous states, voter registries are matters of public record or can be legally purchased; North Carolina, for instance, publishes them online, and parties, campaigns, researchers, and consultants buy them routinely. Possessing a copy does not grant access to the official system, nor does it allow anyone to alter records or vote on behalf of another. A report by the National Intelligence Council (NIC) of October 2020 classified Chinese activity in that cycle as “low-level” and restricted to exploratory steps. Furthermore, the declassified 2021 assessment, compiled under officials appointed by Trump himself, concluded that China considered attempting to influence the outcome but decided against it out of fear of damaging bilateral relations. Conversely, a January 2020 report from the same body supports the president’s general premise: Russia, China, Iran, and North Korea possess the technical capabilities to access US electoral data and potentially manipulate it.
The reaction was instantaneous. The Chinese Embassy in Washington denied all charges using its standard formula, invoking the principle of non-interference. The twenty-four Democratic governors issued a joint statement accusing the president of attempting to intimidate and silence voters. NBC and CNN chose not to broadcast the address. Meanwhile, as Trump enumerated his charges in Washington, Xi Jinping was opening China’s flagship artificial intelligence summit in Shanghai, presenting himself as a responsible global leader in technological governance. The juxtaposition speaks volumes.
Implications
A new battlefield exists, and it is no journalistic metaphor. When a political adversary stands firm against an external enemy that is not a democracy, that external enemy—precisely because it is not one—can resort to methods infinitely more swift and ruthless than those a democracy allows itself. It can employ its intelligence and counter-intelligence services to infiltrate the rival country on a massive scale. This is done not merely to manipulate elections, which is the obvious and least interesting part, but also to manipulate journalism, opinion makers, academia, and culture. The ultimate goal is to drag as many citizens of the adversary nation as possible into distrusting their own government and institutions, leading them to serve the interests of the rival power. The technical term for this, established in 1948, is political warfare.
George Kennan defined it in a Policy Planning Staff memorandum on 30 April of that year: the employment of all means at a nation’s command, short of war, to achieve its national objectives. This is long-standing doctrine, not a modern intuition. In the Chinese case, this doctrine is openly published. Unrestricted Warfare, the 1999 essay by Colonels Qiao Liang and Wang Xiangsui, proclaims the dissolution of the boundary between the military and civil spheres: any arena—financial, media, legal, or cultural—is a theatre of operations. The so-called Three Warfares, incorporated by the Central Military Commission in 2003, comprise public opinion warfare, psychological warfare, and legal warfare (or lawfare), though not in the sense the term is used by the Spanish left.
Russian President Vladimir Putin and Chinese President Xi Jinping have tea after their meeting at the Great Hall of the People in Beijing, China, on 20 May 2026 – SPUTNIK/ALEZANDER KAZAKOV
The United Front Work Department, which Mao termed a “magic weapon” and whose role Xi Jinping has personally reaffirmed, serves as the organic instrument for these efforts. The baseline monograph remains Magic Weapons by Anne-Marie Brady, published by the Wilson Center in 2017 concerning Chinese penetration in New Zealand. This should be required reading for those in Europe who still believe this is merely an American obsession.
The genealogy is Soviet. Active measures (aktivnye meropriyatiya) and disinformation (dezinformatsiya) constituted the very soul of the KGB’s work. The Chinese pattern of mass data aggregation is a matter of record established before courts: the Office of Personnel Management (OPM) in 2015, with over twenty million security clearance files stolen; Anthem in the same year; Equifax in 2017, leading to a formal indictment by the US Department of Justice in 2020 against four officers of the People’s Liberation Army’s 54th Research Institute; and Marriott in 2018. That Beijing has spent a decade assembling a strategic archive on the American population is not a hypothesis; it is a matter of judicial record.
That being said, and this is where our commitment to the truth must outweigh political alignment, the following must be stated clearly: the strategic objective of an influence operation is not to ensure a specific candidate wins; it is to ensure that no result is believed. Thomas Rid demonstrated this in Active Measures (2020) with a thesis that should be framed in every foreign ministry: disinformation does not create fractures, it exploits them; and the domestic reaction to the operation often corrodes trust more than the operation itself. If this is true—and it is—then a presidential denunciation delivered without proof of vote manipulation, four months before the midterm elections, effectively accomplishes the very objective Beijing could not achieve on its own.
We may call this the sentinel’s paradox: the party who denounces infiltration with more noise than evidence ends up executing, at no cost to the adversary, the final phase of the operation they are exposing. It requires no malice; it requires only that partisan interest and the interest of the rival power converge for a brief moment. Robert Chesney and Danielle Citron formulated this in 2019 with an unbeatable phrase: the liar’s dividend. When everything can be fake, the liar always wins. Hannah Arendt foresaw this in The Origins of Totalitarianism (1951): the ideal subject of totalitarian rule is not the convinced fanatic, but the person for whom the distinction between fact and fiction no longer exists. That is the victory Beijing seeks.
We are exceptionally vigilant regarding Chinese expansionism: in the South China Sea, the Pacific, Sri Lanka, the Maldives, Africa, Latin America, and in the control of strategic raw materials and rare earths. For that very reason, we cannot allow the exposure of that threat to degenerate into domestic polarization. A polarized enemy divided into irreconcilable camps is a weak and nearly defeated enemy. When an accusation of this magnitude is instrumentalized, it ceases to serve as a defence and begins to aid the accused.
The collateral damage is something Europe should view with dread: if the president of the world’s leading democracy proclaims in prime time that his electoral system is comparable to that of the third world, with what moral authority can the West demand clean elections in Caracas, Managua, or Minsk? The Rodríguez siblings, Delcy and Jorge, who inherited from Maduro the largest mafia organisation ever disguised as a state, took note last night. And they were not alone.
Perspectives and Scenarios
First Scenario (Inertia): The Republican-majority Congress fails to pass the Save America Act, the matter fades into the noise, and the declassified material remains as stored political ammunition.
Second Scenario (The November Mortgage): If Republicans lose the midterms, the material declassified tonight becomes the evidentiary framework for a formal challenge. This is an insurance policy against defeat.
Third Scenario (The Chinese Response): Beijing needs to do nothing. It is enough to do what it did last night: offer a formal diplomatic denial while simultaneously appearing in Shanghai as a responsible guarantor of the global technological order.
Fourth Scenario (Absent Europe): Our political class will remain oblivious. European voter registries, health records, and administrative databases are just as vulnerable. Europe has spent twenty years delegating its informational sovereignty. This is pure dereliction of duty.
Xi Jinping’s lunar programme progresses with firm and methodical steps, with the intention of Chinese astronauts setting foot on the Moon by 2030 or earlier. In image, the president at the Kremlin in May of the current year – PHOTO/Kremlin
Unemployed engineers, hired philosophers: the moral restructuring of artificial intelligence
The Facts
Until very recently, software engineers were the most sought-after professionals on the planet. Artificial intelligence companies poached the best talent from one another with signing bonuses reaching five, six, and seven figures before the employee had even logged a single hour of work. Then, suddenly, the collapse arrived.
According to data from the Federal Reserve Bank of New York on the labour market for recent graduates (aged 22 to 27), recent software engineering graduates face an unemployment rate of around 7.5%, and computer science graduates face roughly 6.1%, compared to 5.7% for recent university graduates as a whole. It is worth noting that this figure refers specifically to recent graduates and not the entire profession: veterans are not included in this number, and the US Bureau of Labor Statistics (BLS) continues to project 15% job growth for software developers between 2024 and 2034. However, the entry gate has slammed shut: junior generalist job openings have plummeted by roughly 30% year-on-year. Computer science enrollment fell by 8.1% in the 2025–2026 academic year, the sharpest decline of any major.
Meanwhile, a phenomenon has emerged that few saw coming: while engineers stand in line, major artificial intelligence laboratories are hiring philosophers. This is a documented trend. Leading frontier companies such as Anthropic and Google DeepMind are paying professional philosophers between $250,000 and $400,000 annually to translate traditional moral systems into constitutional rules that govern model behavior. The names are public: Amanda Askell serves as the philosopher-in-residence at Anthropic, alongside Joe Carlsmith, Ben Levinstein, Jackson Kernion, and Harvey Lederman. Iason Gabriel and Henry Shevlin are embedded at Google DeepMind. Philosophers act as ethical gatekeepers, determining what information, values, and modes of reasoning should shape these systems.
A detail that confirms this dynamic: one of the core principles these laboratories employ in model training is Socratic ignorance, used specifically to combat sycophancy—the model’s tendency to tell users exactly what they want to hear. Maieutics is operating not as an ornamental metaphor, but as an engineering technique. Socrates is on the payroll.
Just yesterday, The Washington Times published a counter-perspective: labour market experts tempered the excitement, pointing out that only seven to eight thousand students graduate with a philosophy degree each year in the United States, meaning these reports could create false expectations of a generalized boom that does not exist. The most uncomfortable objection remains that philosophical research risks becoming an extension of marketing departments, as those signing the payroll ultimately answer to investors and shareholders.
Morocco will use blockchain technology for the development of the Digital Morocco 2030 Plan – Depositphotos
Implications
The hiring of philosophers stems from practical necessity. Socratic maieutics allows models to question themselves and learn beyond the explicit prompt provided by the user. However, beneath the learning methodologies lie essential ethical and moral questions. And that is precisely where the problem begins.
The core issue is that certain companies are trying to rely on a single philosophical school, or two at most, because their operational scope is narrow. This is an error of immeasurable proportions. Under no circumstances should a general-purpose artificial intelligence model—and even less so one deployed in defence or autonomous driving—rely exclusively on consequentialism.
Fiction illustrated this perfectly in the film I, Robot (2004): a robot rescues an adult while allowing a twelve-year-old girl to drown because it calculated that the adult had a 45% chance of survival compared to her 11%. The protagonist’s response cuts to the heart of the moral dilemma: she was someone’s daughter; eleven percent is more than enough; a human being would have known. That is pure consequentialism. Transfer this to defence and geopolitics, and imagine pure consequentialism in an autonomous weapon. It will select military targets based entirely on a cold optimization of collateral variables. No moral conscience. No mercy. Just arithmetic. Once human life is reduced to an optimization variable, the question is no longer if the machine will fail, but when and at what human cost. Let us not even speak of finalism (the end justifying the means), which would represent the ultimate disaster for humanity.
Thus, it is imperative to feed artificial intelligence systems with all branches of philosophy, morality, and ethics. We must incorporate Kantian deontology: the categorical imperative, which mandates treating humanity always as an end and never merely as a means. We must look to Hegel and the classics. And we must also turn to philosophers with religious roots, whose moral and ethical frameworks are exceptionally strict and clear. In this regard, I would look directly to St Augustine, the father of just war theory, who brings something consequentialism can never provide: the ordo amoris—the order of love, the conviction that an objective hierarchy of goods exists that is recognized, not calculated.
This brings us to the encyclical, a document almost entirely ignored in Europe despite being the most important text published on this matter this century. Magnifica Humanitas, “On the Safeguarding of the Human Person in the Era of Artificial Intelligence”, is a monumental milestone. The text was presented at the Vatican alongside major developers, who acknowledged before the official audience of the Holy See that software engineers alone cannot map out the ethical boundaries of AI due to systemic market and competitive pressures.
The encyclical’s premise is remarkably sober: technology is not an antagonistic force to humanity, nor is it inherently evil; however, technology is never neutral. Hence the central warning of paragraph 107: it is not enough to create a more moral AI if that morality is dictated by a select few. The decisive passage, paragraph 199, stipulates that no algorithm can render war morally acceptable; artificial intelligence does not eliminate the intrinsic inhumanity of conflict but risks accelerating it and making it more impersonal, reducing victims to mere data points and optimizing violence. Any technology that facilitates striking an enemy without seeing their face lowers the moral threshold of conflict.
Logo of the alphabet and the words of Artificial Intelligence (AI) – REUTERS/DADO RUVIC
Perspectives and Scenarios
First Scenario (Capture): The most probable and dangerous outcome. Philosophers enter corporate laboratories and their role degrades into providing moral rubber-stamping or “ethics washing” for pre-determined product decisions.
Second Scenario (The Moral Monopoly): Two or three companies within a single country, drawing predominantly from a single tradition—Anglo-Saxon analytical philosophy—write the moral rules for the entire world, entirely bypassing Thomism, Augustinianism, or continental traditions.
Third Scenario (The Military Bifurcation): This crucial debate is entirely absent where it matters most: within autonomous weapons programmes. In lethal autonomous systems, there is no philosopher-in-residence, because there consequentialism is the native language of the targeting system.
Fourth Scenario (Europe’s Absence): While Washington hires philosophers and Beijing hosts global summits, Europe merely regulates. Regulating without building is the most elegant form of irrelevance. It represents a massive cultural capitulation for the continent that birthed international law.
Fifth Scenario (The Restructuring): The market has discovered that AI’s true bottleneck is moral, not technical. The humanities reclaim their vital function. Socrates has returned through the front door.
Editorial commentary
Today’s two major stories are the exact same news told twice.
In Washington, a president denounces an authoritarian power for harvesting the data of 220 million citizens to track who they are and how they vote. The denunciation contains substance in its facts but fails entirely in its conclusion. Political warfare is a reality, and the Chinese pattern of mass data gathering is a judicially proven fact; however, data vulnerability does not automatically equate to vote manipulation. When an accusation of this gravity is weaponized for partisan advantage months before an election, it ceases to serve as a shield and begins to aid the accused. The price of this outburst is paid by the moral credibility of the entire West when demanding clean elections in Caracas or Minsk.
In California, meanwhile, the other half of the same war is being fought in absolute silence. A handful of corporate boardrooms are deciding what morality will govern the machines that will very soon select military targets. Pure consequentialism in an autonomous weapon is not an academic hypothesis; it is the cold math that reduces human life to an optimization variable. In contrast, the Kantian categorical imperative and Augustinian thought remind us that right intention and human dignity matter just as much as the final output.
That a papal encyclical stands as this century’s most profound document on autonomous weapons, while European parliaments debate regulations in a purely bureaucratic fashion, is both hopeful and humiliating. Europe has failed to place a single thinker at the table where the morality of global systems is being written. Let us, however, give credit to Anthropic, which is by far the laboratory taking constitutional ethics and structural alignment most seriously. But no entity can be judge in its own cause: the only true guarantee lies in the existence of well-funded, independent, external critics. Otherwise, the morality written into a machine, once deployed at a global scale, will be impossible for anyone to correct.