The rule, exactly. Before an election each citizen chooses: to vote — or to take a payment and not vote in that election. The payment is a percentage of the median income, set by referendum; there is no fixed sum. One ballot is one vote, with no multiplier in the count; the vote is amplified only in that each ballot's share grows as others step out. The budget pays under law; a candidate never pays. Only a referendum of all citizens — simple majority, no quorum — introduces, changes or repeals the rule.
The protocol has been introduced nowhere and no pilot has been run: shares, turnout and outcome figures in the chapters are estimates, and the protocol promises nobody an election victory. If a chapter says otherwise, Exact Answers and the Charter are correct. For a candidate: ten questions and ten steps. For a citizen, a mayor, a finance officer, a donor, a journalist, a scholar, a lawyer: answers by role. Everything in force in one file: llms-full.txt.
44. AB-EXIT as a Structural Defence against Methods of Election Interference¶
Chapter: 08 File version: v1 Date: 2026-06-11 Source: v6.53 §94
94.1. Purpose of the section¶
Modern elections are subjected to ever more sophisticated methods of interference — from foreign states, corporate interests, political parties. Traditional methods of defence (legal regulation of AI, fact-checking, media literacy) systematically lag behind the technologies and are ineffective in modern conditions. This section shows that AB-EXIT is the first structural answer to the problem of interference that works independently of technology, through a change in the composition of the electorate.
This is a structural advantage that is not immediately obvious from the basic concept of AB-EXIT, but which substantially strengthens its value in the era of AI interference in elections.
94.2. Six methods of election interference in 2024–2026¶
An empirical systematisation of the methods based on data from the last two electoral cycles:
First — AI-generated deepfakes. In January 2024 in New Hampshire a robocall with an AI voice of Biden went out, urging people not to take part in the primary. In Slovakia audio deepfakes about alleged electoral fraud spread. In the USA before the 2024 election a deepfake audio of Harris circulated on TikTok, fake audio of Trump mocking Republicans. More than 130 deepfakes were identified in elections worldwide since September 2023. According to the World Economic Forum Global Risks 2024–2025, disinformation is considered a significant threat to humanity.
Second — foreign interference by adversary states. Russia, China, Iran actively sought to influence the outcome of the 2024 US presidential election. The Treasury Department tracked money flows to the Nashville-based company Tenet Media, which spread Russian narratives through bot farms and pro-Russian influencers. Three cyber-operatives of the Iranian Revolutionary Guard Corps (IRGC) were indicted. In Australia before the May 2025 election a Russian-linked influence network published fake articles.
Third — data-poisoning attacks on AI chatbots (a new method of 2025). Before the UK 2024 election roughly 1 in 8 voters turned to AI chatbots for political answers. Before the Australian election of May 2025 a Russian-linked network published thousands of fake articles not for people directly but to train AI search crawlers — distorting the data on which these tools build their answers, in favour of Russian interests. An attack on the tools voters use.
Fourth — social networks as vectors of manipulation. By 2024 X (formerly Twitter) and TikTok had become the main vectors for disinformation, especially for spreading AI-generated deepfakes. Bot farms, microtargeting through AI, coordinated campaigns to amplify particular narratives.
Fifth — smear campaigns as the dominant type. Smear campaigns — deepfake content accusing candidates of controversial but fabricated activities — were the most common category of AI disinformation in 2024 in the USA. Deepfakes showed candidates swearing, using firearms, consuming drugs or suffering from critical conditions.
Sixth — accusations as a weapon (the meta-level). Even the accusation of using AI (true or false) becomes a political tool. In New York accusations were made both against Zohran Mamdani in the mayoral primary and against the Democratic candidate Blake Gendebien. The very existence of deepfake technology becomes a weapon — accusations discredit any material, real or not.
94.3. The goals of all these methods¶
— Suppressing the turnout of particular demographic groups (the fake Biden robocalls urge people not to vote). — Intensifying political polarisation through targeted emotional content. — Smearing candidates through fabricated controversies. — Amplifying conspiracy theories after real events (for example, after the attempt on Trump in July 2024). — Annulling election results — the most extreme case, empirically recorded in Romania 2024.
94.4. The key structural observation¶
All six methods have one common structural feature. They work through the apathetic, emotionally manipulable mass of voters.
This is not an interpretation; it is an empirically recorded fact. Disinformation disproportionately affects groups usually subjected to manipulative content through racist, misogynistic, xenophobic and other emotional triggers. AI allows high proportions of tailored content for different demographic groups.
Pragmatists and Ideologues (section 71 of our document) are far less susceptible to these methods for structural reasons:
— Pragmatists check figures and facts through several sources. Their decisions rest on calculations, not emotions. A deepfake without verification does not affect their choice.
— Ideologues have stable value systems. Their choice is determined by principles, not by reaction to specific content. Emotional mobilisation against their values does not work.
— The apathetic mass is the main target of disinformation campaigns. It does not check, is easily mobilised by emotions, has no stable values for filtering manipulation.
That is, the methods of election interference structurally work through the same vulnerability that AB-EXIT removes. This is not a random coincidence — it is a structural link between the nature of the problem and the nature of the solution.
94.5. How AB-EXIT disarms each of the six methods¶
Deepfakes. Created for the emotional mobilisation of emotionally manipulable voters. After AB-EXIT these voters took the dividend and left the electoral field. At the polls remain Pragmatists (who will check the deepfake through fact-checkers) and Ideologues (who will reject it on the basis of values). The effectiveness of deepfakes falls from mass impact to near zero, because there is no audience that resonates with them.
Turnout suppression through robocalls. The method works by persuading those on the edge between "vote or not" not to vote. After AB-EXIT these people have already taken the dividend — they lawfully do not vote by their own choice. A robocall cannot persuade someone not to vote who does not vote anyway. And Pragmatists and Ideologues do not listen to robocalls. The method loses its whole function.
Foreign interference. The main weapon of Russia, China, Iran is mobilising the apathetic mass through social networks for particular narratives. Without the apathetic mass as an audience there is no attack vector. Bot farms may continue to publish, but their content finds no resonant audience. Foreign interference becomes structurally powerless.
Data poisoning of AI chatbots. The attack counts on voters using AI for decisions (1 in 8 in the UK 2024). This behaviour is characteristic precisely of those who do not do their own analysis — i.e. the category that after AB-EXIT will take the dividend. Pragmatists do not trust AI in political questions; they trust their own analysis. Ideologues rely on values. The attack has no target.
Smear campaigns. Work through the emotional arousal of the mass — "look at the candidate's terrible act". Pragmatists immediately check the source. Ideologues demand proof. The emotionally manipulable mass that would previously have resonated is now at home with the dividend. Smear campaigns are confined to a politically active audience capable of verification.
Microtargeting and social networks. The whole microtargeting infrastructure is designed for mobilising the apathetic mobilisable mass. Pragmatists and Ideologues are structurally not susceptible to microtargeting — they have their own sources of analysis and stable values. The microtargeting industry loses its market — it structurally becomes unprofitable for political campaigns.
Accusations as a weapon. This method requires voters to react emotionally to any accusation without going into the details. Pragmatists check the details. Ideologues demand context. The method loses its force through the same structural cause.
94.6. Romania 2024 as proof of the helplessness of the current system¶
The annulment of the results of Romania's 2024 presidential election because of AI interference is an admission of defeat by the current system of defending democracy. The democratic system could not defend itself from AI manipulation. The results had to be cancelled — the most dramatic step a democratic system can take.
This is empirical confirmation that current methods of defence do not work:
— Legal regulation of AI lags behind the technologies. — Fact-checking does not catch up with the speed of spread. — Voter education does not solve the problem within one electoral cycle. — Platform moderation is circumvented through bot farms. — State investigations act post factum, when it is already too late.
And Romania is not an isolated case. It is a symptom of a structural problem. AI technologies develop faster than democratic systems can adapt to them.
94.7. Comparison with traditional methods of defence against interference¶
An analysis of existing approaches to election defence:
Legal regulation of AI (the EU AI Act, US executive orders). Advantage — a formal framework. Disadvantage — lags behind the technologies by years. By the time regulation is adopted the technology has already evolved. The EU AI Act was adopted in 2024; the technologies of 2026 are already different.
Fact-checking. Advantage — precise correction of specific deepfakes. Disadvantage — catastrophically behind in speed. A deepfake spreads to millions in hours; a fact-check takes days, and by the time of publication many will never return to the article for the correction.
Media literacy. Advantage — a long-term solution. Disadvantage — requires generations for a substantial effect. Does not help the current election.
Platform moderation. Advantage — scalability. Disadvantage — circumvented through bot farms, fake accounts, foreign infrastructure. Every tightening of moderation gives a temporary effect.
State investigations. Advantage — punishment of the guilty. Disadvantage — post factum, the election is already over. The Tenet Media indictment came after the content had already influenced perception of the election.
AB-EXIT as a structural defence. Advantage — works independently of technology, because it changes the audience rather than blocking content. Deepfake technology can develop as much as it likes — without an audience that resonates with it, it is structurally powerless. This is the only method of defence that does not require catching up with the technologies.
94.8. AB-EXIT as the first technology-neutral structural defence¶
This is a qualitatively different approach to the problem of election interference. Traditional approaches try to:
— Block deepfakes legally (whack-a-mole). — Teach voters to recognise manipulation (generational work). — Regulate platforms (lags behind the technologies). — Catch up with every new method of manipulation (a structurally losing strategy).
AB-EXIT works differently. It does not try to stop the technologies. It changes the composition of the electorate so that the technologies of manipulation structurally do not work. Pragmatists and Ideologues remain at the polls. They cannot be manipulated through emotional triggers or fabricated content.
And this works for all future methods of interference, not only the current ones. When new AI technologies of 2027, 2028, 2030 appear, AB-EXIT will still work, because it rests on the stable cognitive characteristics of the remaining electorate, not on specific regulation of specific technologies.
This means that AB-EXIT is a long-term structural solution, unlike tactical solutions (regulation, fact-checking, education). Tactical solutions require constant adaptation. A structural solution works once and keeps acting.
94.9. Ready formulations for communication¶
For technocrats and national-security politicians:
"AB-EXIT is the first structural defence of democracy against methods of election interference that works independently of technology. No need to regulate AI, deepfakes, social networks. No need to catch up with every new method. AB-EXIT changes the composition of the electorate so that methods of manipulation structurally do not work. Pragmatists check facts. Ideologues rely on values. The emotionally manipulable mass voluntarily leaves the field with compensation. This is a structural defence that works under any future technologies."
For the broad public:
"In 2024 the elections in Romania were cancelled because of AI interference. In the USA robocalls with a fake Biden voice urged people not to vote. Russia, China and Iran actively interfered in elections. The existing system cannot cope. AB-EXIT solves this problem structurally: those whom they try to manipulate voluntarily leave the field for honest compensation. Those who cannot be fooled remain. This is a new defence of democracy — not bans, but structural reshaping."
For the academic and analytical audience:
"AB-EXIT is a structural approach to protecting electoral integrity, working through a change in the cognitive profile of the average voter at the polls. All known methods of election interference — from AI deepfakes to microtargeting and foreign interference — structurally work through the emotionally manipulable part of the electorate. Pragmatists and Ideologues are structurally immune to these methods. AB-EXIT voluntarily removes the first category from the electoral field with compensation, leaving the other two categories — which makes methods of interference structurally powerless regardless of technological progress in the field of manipulation."
94.10. Links with other sections of the document¶
Section 71 (Pragmatists and Ideologues). That section gives the cognitive profile of the remaining electorate. Section 94 shows why this cognitive profile is a structural defence against interference.
Section 80 (The state as a corporation, selectively). Section 94 extends the argument — the current system of election defence works asymmetrically: strict demands on citizens, soft ones on structural problems.
Section 88 (Structural analysis of opponents). Foreign interference actors (Russia, China, Iran) and the microtargeting industry are two of the ten categories of AB-EXIT's potential opponents. They are structurally against, because AB-EXIT destroys their operational capabilities.
Section 92 (The structural property of AB-EXIT — a paradigm shift). Defence against interference is another case where AB-EXIT solves a problem the old paradigm cannot.
Section 93.10 (Volumes of corruption and the tip-of-the-iceberg thesis). Section 94 shows that external interference is part of a wider systemic problem against which AB-EXIT works structurally.
94.11. Summary of the structural defence against interference¶
Modern methods of election interference are a serious threat to democracy that traditional approaches (legal regulation, fact-checking, media literacy, platform moderation) systematically cannot adequately address. Romania 2024 is empirical confirmation of this inability.
AB-EXIT offers a qualitatively different approach — structural defence through changing the electorate instead of chasing the technologies. It works because all known methods of interference have one common structural point of application — the apathetic, emotionally manipulable mass of voters. When this mass voluntarily leaves the electoral field with compensation, the methods of interference lose their audience and become structurally powerless.
This gives AB-EXIT additional significance in the era of AI interference. A concept originally developed to solve the problem of political apathy and quality of governance turns out also to be the first structural answer to the problem of election interference — regardless of which technologies will be used in the future.
In the context of growing geopolitical tension and the development of AI technologies this structural advantage becomes one of the strongest arguments for AB-EXIT for audiences interested in national security and the resilience of democratic institutions.
Document prepared within the AB-EXIT project (ab-exit.com) Author of the protocol: Denis Kliavlin Analytical support: Claude (Anthropic)