Seedlabs

Demagogic Rhetoric Pattern Scanner

A tool for newsrooms that scans political speeches for patterns of leader-worship and aggressive rhetoric.

An API that scores political speeches, ads, and channels for the four psychological signatures of demagogic propaganda — leader-worship framing, slogan repetition without policy content, out-group aggression, and regression-inducing simplification — giving newsrooms and ad-verification firms a structural read that fact-checking cannot provide.

PsychologyPsychoanalysis, Philosophy, and Politics
Career Development and DiversityNewsroom fact-checking and ad verification (media intelligence / trust & safety)

Concept

A text- and transcript-scoring API that audits political content — speeches, campaign ads, cable segments, influencer channels — for the four psychological signatures of demagogic propaganda identified in the paper: (1) ego-ideal staging: omnipotent-authority and savior framing built around a leader; (2) repetition density: slogan recycling relative to rational policy content; (3) out-group aggression: named enemies and the emotional aggression channeled toward them; (4) regression cues: instinctual, crowd-addressing simplification. The output is a per-message 'rhetoric mechanics' score plus trend lines per speaker and channel, with alerts when a channel's mix shifts (e.g., aggression targets multiplying). The wedge: it is a structure checker, not a fact-checker — it scores how a message works psychologically, not whether its claims are true. Sold as an API and dashboard to fact-check desks, extremism monitors, and ad-verification firms.

Why now

The paper demonstrates that fascist propaganda does not persuade through 'the rational statement of rational aims' but relies 'overwhelmingly on the repetitive mobilization of unconscious psychological mechanisms' — staging the leader as 'an object of narcissistic identification and omnipotent authority,' driving 'artificial regression' through methodical repetition, and manufacturing 'irrational emotional aggressiveness' toward designated out-groups. These are enumerable, countable structural features (repetition rates, authority framing, named targets) — exactly what automated text analysis can measure and exactly what claim-level fact-checking structurally misses. The corpus's hegemony framework [4] independently establishes that mass politics is organized through collective identification rather than rational interest, reinforcing that mechanism-level detection is the right layer of analysis. Readiness caveat: the paper is a theoretical frame, not a validated classifier — the product requires an annotated corpus, human-in-the-loop review, and scores framed as rhetorical-pattern measures rather than verdicts.

Written by GLM · 11 Oct 2026

AI assessment

Backed by 2 papers76

A high-potential structural analysis tool that pivots from 'truth-checking' to 'mechanism-checking,' though it faces significant technical hurdles in translating 1950s psychoanalytic theory into a reliable ML classifier.

Evidence strength
3/5
The idea is grounded in a strong theoretical framework by Adorno, but the research is qualitative and psychoanalytic rather than a quantitative dataset, meaning the 'signatures' are not yet computationally defined.
Market pull
4/5
There is high urgency and existing budget among trust & safety organizations and newsrooms to identify systemic disinformation patterns that bypass traditional fact-checking.
Novelty & moat
5/5
Unlike the listed catalog items which focus on politeness or authenticity, this introduces a novel psychological-structural approach to political discourse analysis.
Feasibility
2/5
Building a classifier for 'artificial regression' and 'ego-ideal staging' requires a massive, expertly annotated corpus that does not currently exist, making the MVP high-risk.
Wedge clarity
4/5
The distinction between 'structure' and 'fact' provides a sharp, defensible entry point that avoids the political minefield of deciding what is 'true'.
Simplicity / focus
5/5
The product is a single, focused API with a clear output (a rhetoric score), avoiding the trap of becoming a broad 'democracy platform'.

Scored by Gemma 4 · 11 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.

Persona discussion

AI personas trained on real people's expertise debate this idea as it evolves.

View the discussion →

Act on this idea

Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.

Who benefits

  • Institute for Strategic Dialogueorganization

    ISD's analysts track extremist propaganda across platforms; the scanner gives them a repeatable structural taxonomy — leader-identification, repetition, out-group aggression — to quantify escalation trends instead of relying on qualitative coding.

  • Anti-Defamation Leagueorganization

    ADL monitors hate and extremism trends; structural scoring of out-group aggression escalation across channels directly supports their public trend reporting and incident analysis.

  • Zefrcompany

    Zefr classifies video content for brand safety; a demagogic-pattern score catches leader-cult and out-group aggression content that keyword blocklists miss, keeping advertiser spend off monetized aggression.

  • NewsGuardcompany

    NewsGuard rates information sources; a rhetoric-mechanics layer complements its fact-based ratings with a how-it-persuades dimension for its newsroom and platform clients.

Research it builds on

  1. Hegemony and Socialist Strategy: Towards a Radical Democratic Politics
    Chantal Mouffe, Ernesto Laclau · 1985 · 9313 citations
    All ideas from this paper →
  2. Freudian Theory and the Pattern of Fascist Propaganda
    Theodor Wiesengrund Adorno · 2026 · 109 citations
    All ideas from this paper →
Leave feedback
feasibility