Puppets or partners? Governing cyborg propaganda in the digital public square

Anton Gollwitzer; Asbj{\o}rn F{\o}lstad; Daniel Thilo Schroeder; Gary Marcus; Jay J. Van Bavel; Jonas R. Kunst; Jon Roozenbeek; Kinga Bierwiaczonek; Live Leonhardsen Wilhelmsen; Marc Fawcett-Atkinson

arxiv: 2602.13088 · v2 · submitted 2026-02-13 · 💻 cs.CY · cs.AI

Puppets or partners? Governing cyborg propaganda in the digital public square

Jonas R. Kunst , Kinga Bierwiaczonek , Meeyoung Cha , Omid V. Ebrahimi , Marc Fawcett-Atkinson , Asbj{\o}rn F{\o}lstad , Anton Gollwitzer , Nils K\"obis

show 7 more authors

Gary Marcus Jon Roozenbeek Daniel Thilo Schroeder Jay J. Van Bavel Sander van der Linden Rory White Live Leonhardsen Wilhelmsen

This is my paper

Pith reviewed 2026-05-15 22:17 UTC · model grok-4.3

classification 💻 cs.CY cs.AI

keywords cyborg propagandaalgorithmic influencepolitical discourseregulatory frameworkshuman-bot binarydigital governanceinfluence operationsdemocratic discourse

0 comments

The pith

Cyborg propaganda pairs verified human accounts with algorithmic automation to generate personalized content at scale and shift political discourse into algorithmic campaigns.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper develops cyborg propaganda as a hybrid system that links real citizen accounts to automated tools for creating and distributing tailored political messages. It shows how this setup exploits gaps in regulations that assume a strict separation between human activity and bot behavior, including measures like the EU AI Act. A reader would care because the result turns open debate over competing ideas into contests between hidden algorithmic operations, with ordinary users serving as cover for coordinated influence. The analysis compares responses across democratic and non-democratic settings and identifies three concrete regulatory steps to address the problem at micro, meso, and macro levels.

Core claim

The authors define cyborg propaganda as a closed-loop architecture that combines verified human accounts with algorithmic automation to produce personalized content at scale. This hybrid form exploits regulatory gray zones created by frameworks that rely on the human/bot binary, allowing campaigns to ratify AI-generated messages through real citizens. The result shifts political discourse from a contest of ideas to a battle of algorithmic campaigns, reducing citizens to cognitive proxies. Comparative examination reveals that democratic states are most capable of regulation yet most constrained by rule-of-law requirements, while non-democratic actors face no such limits, making international

What carries the argument

Cyborg propaganda, defined as a closed-loop architecture combining verified human accounts with algorithmic automation to generate personalized content at scale, which operates in the regulatory gap between human and bot categories.

If this is right

Classifying coordination platforms as political action committees would require supply-chain transparency for these hybrid campaigns.
Mandating researcher access to platform data through DSA-style mechanisms would allow monitoring of synthetically coordinated activity.
Risk standards that penalize amplification of synthetically coordinated content would create deterrence across platforms.
Democratic states hold the greatest regulatory capacity but face the strongest rule-of-law limits, while non-democratic actors lack internal accountability.
International risk standards become the main cross-border enforcement tool because domestic approaches cannot reach non-democratic operators.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Platforms may need to build new detection methods that flag coordinated patterns across human and AI elements rather than relying on bot classifiers alone.
The framework could extend to commercial or public-health messaging where similar hybrid coordination occurs at scale.
Empirical tracking of message origins from accounts showing mixed human and automated signals could quantify how much discourse has already shifted to algorithmic campaigns.
Citizens might require new verification tools to distinguish genuine individual posts from those routed through coordination hubs.

Load-bearing premise

Frameworks built on a strict human/bot binary are structurally unable to regulate hybrid systems that combine both elements.

What would settle it

A successful demonstration that existing human-bot detection tools or laws can identify and limit large-scale cyborg propaganda campaigns without new classifications or data-access rules would show the binary frameworks are adequate.

read the original abstract

The distinction between genuine grassroots activism and automated influence operations is collapsing. While contemporary policy debates prioritize fully autonomous generative agents and synthetic content, this paper offers a conceptual contribution: we develop 'cyborg propaganda,' a closed-loop architecture combining verified human accounts with algorithmic automation to generate personalized content at scale, as a distinct and undertheorized threat to democratic discourse. By relying on verified citizens to ratify AI-generated messages, these campaigns exploit a regulatory gray zone that frameworks built on the human/bot binary (including the EU AI Act and Section 230) are structurally unable to address. Drawing on a conceptual analysis of coordination platforms and comparative examination of governance frameworks across democratic and non-democratic contexts, we analyze this paradox across micro, meso, and macro levels. We examine whether cyborg propaganda democratizes political power by unionizing influence or reduces citizens to cognitive proxies of a hidden directive, arguing that it shifts political discourse from a contest of ideas to a battle of algorithmic campaigns. We propose three regulatory responses: classifying coordination hubs as political action committees to enforce supply-chain transparency; mandating researcher access to platform data through DSA-style mechanisms; and establishing risk standards penalizing amplification of synthetically coordinated content. Comparative analysis reveals that viability varies structurally. Democratic states are simultaneously the most capable of regulation and the most rule-of-law constrained. By contrast, non-democratic actors face no comparable accountability, making international risk standards the primary cross-border enforcement mechanism.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper names 'cyborg propaganda' as verified-human-plus-AI campaigns and flags a possible regulatory gap, but the gap claim rests on assertion rather than statutory detail or cases.

read the letter

The central move is to define cyborg propaganda as a closed loop where verified accounts ratify AI-generated political content at scale. This is presented as distinct from pure bots or pure grassroots, and the authors map it across micro, meso, and macro levels while comparing democratic and non-democratic governance contexts. The three concrete proposals—treating coordination platforms as PACs, requiring DSA-style data access for researchers, and setting amplification risk standards—give readers something specific to evaluate. The observation that non-democratic actors face fewer internal constraints is straightforward and useful for thinking about cross-border effects. Those parts are clear and structured enough to discuss in a policy seminar. The main limitation is that the regulatory-gap argument stays at the level of logic. The paper states that human/bot binaries in the EU AI Act and Section 230 cannot reach these hybrids, yet it supplies no close reading of the relevant statutory language, no enforcement examples, and no evidence on how often verified humans actually front such campaigns. Without that grounding, the claim that discourse shifts from ideas to algorithmic battles remains untested. This is a conceptual policy piece aimed at readers working on AI governance, disinformation rules, and platform regulation. Someone already following the EU AI Act or Section 230 updates would find the framing worth considering, but anyone wanting data, prevalence estimates, or falsifiable tests would not. I would send it to peer review. The definitional contribution is timely and the comparative structure is workable, but referees would need to press on the legal and empirical thin spots before the regulatory recommendations could be taken as more than starting points.

Referee Report

2 major / 2 minor

Summary. The paper introduces 'cyborg propaganda' as a closed-loop hybrid architecture in which verified human accounts ratify and amplify AI-generated personalized political content at scale. It claims this exploits a regulatory gray zone because existing frameworks (EU AI Act, Section 230) rest on a human/bot binary and are therefore structurally unable to reach such systems. Through conceptual analysis of coordination platforms and comparative review of governance regimes across democratic and non-democratic contexts, the manuscript examines the phenomenon at micro, meso, and macro levels, argues that it converts political discourse from a contest of ideas into a battle of algorithmic campaigns, and proposes three regulatory responses: classifying coordination hubs as political action committees, mandating DSA-style researcher data access, and establishing risk standards for synthetically coordinated amplification.

Significance. If the definitional distinction and the structural-inability claim hold, the paper supplies a useful conceptual lens for an emerging class of hybrid influence operations and identifies regime-type differences in regulatory capacity. Its comparative framing across democratic and non-democratic settings is a clear strength, as is the explicit linkage of micro-level architecture to macro-level governance proposals. The absence of empirical prevalence data or falsifiable tests, however, confines the contribution to the conceptual and normative domain.

major comments (2)

[Abstract and governance frameworks section] Abstract and the section on governance frameworks: the central assertion that frameworks built on the human/bot binary are 'structurally unable to address' cyborg propaganda is stated without a close reading of specific statutory language (e.g., EU AI Act transparency or high-risk obligations, or Section 230 safe-harbor provisions) or any mapping of the closed-loop architecture onto existing disclosure or liability rules. This gap is load-bearing for the regulatory-gap claim.
[Abstract and comparative analysis section] Abstract and the comparative analysis section: the argument that cyborg propaganda shifts discourse 'from a contest of ideas to a battle of algorithmic campaigns' rests on the premise that verified-human ratification of AI content evades current accountability mechanisms, yet the manuscript supplies no enforcement examples or case studies showing why platform detection of coordinated inauthentic behavior or human-liability doctrines would fail to reach these campaigns.

minor comments (2)

[Regulatory responses section] The three proposed regulatory responses are presented at a high level; adding even brief discussion of implementation feasibility (e.g., how 'supply-chain transparency' would be operationalized for decentralized platforms) would strengthen the policy section.
[Analytical framework section] The micro/meso/macro analytical structure is announced but the transitions between levels are not always explicit; clearer signposting would improve readability.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive feedback, which identifies important gaps in the substantiation of our regulatory-gap claims. We respond to each major comment below and outline the revisions we will undertake.

read point-by-point responses

Referee: Abstract and the section on governance frameworks: the central assertion that frameworks built on the human/bot binary are 'structurally unable to address' cyborg propaganda is stated without a close reading of specific statutory language (e.g., EU AI Act transparency or high-risk obligations, or Section 230 safe-harbor provisions) or any mapping of the closed-loop architecture onto existing disclosure or liability rules. This gap is load-bearing for the regulatory-gap claim.

Authors: We agree that the claim requires more granular statutory analysis. In the revised manuscript we will expand the governance frameworks section with a dedicated subsection that closely reads key provisions of the EU AI Act (transparency obligations for generative AI and high-risk system rules) and Section 230 safe-harbor language. We will explicitly map the closed-loop architecture—verified human ratification of AI-generated content—onto these rules to show why the hybrid structure evades both disclosure triggers and liability standards that presuppose either fully human or fully automated actors. revision: yes
Referee: Abstract and the comparative analysis section: the argument that cyborg propaganda shifts discourse 'from a contest of ideas to a battle of algorithmic campaigns' rests on the premise that verified-human ratification of AI content evades current accountability mechanisms, yet the manuscript supplies no enforcement examples or case studies showing why platform detection of coordinated inauthentic behavior or human-liability doctrines would fail to reach these campaigns.

Authors: We accept that illustrative examples would strengthen the argument. We will add a short subsection with hypothetical enforcement scenarios and references to documented hybrid influence operations in the literature to demonstrate why coordinated inauthentic behavior detection (which targets bot-like patterns) and human-liability doctrines are likely to miss verified-human amplification of synthetic content. We maintain that the core claim remains structural rather than empirical. revision: partial

standing simulated objections not resolved

The request for concrete enforcement examples or case studies of mechanism failure, which would require original empirical data collection beyond the conceptual scope of the manuscript.

Circularity Check

0 steps flagged

No derivation chain; conceptual analysis is self-contained

full rationale

The manuscript is a conceptual and policy-oriented paper that introduces the term 'cyborg propaganda' via definition and then performs comparative analysis of governance frameworks. No equations, fitted parameters, or predictive derivations appear. The core argument—that existing human/bot binary frameworks are structurally unable to address verified-human + AI hybrids—rests on logical assertion and high-level policy comparison rather than any step that reduces to its own inputs by construction. Any self-citations present are incidental and non-load-bearing; the central contribution is definitional and analytical, not a closed mathematical or empirical loop.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

The central claim rests on the assumption that hybrid human-AI systems create a regulatory gray zone not covered by existing human/bot frameworks; no free parameters or invented physical entities, but the new conceptual category itself functions as an invented analytical entity without independent empirical grounding.

axioms (2)

domain assumption Contemporary policy debates prioritize fully autonomous generative agents and synthetic content over hybrid systems
Stated directly in the abstract as the motivation for the contribution
domain assumption Frameworks built on the human/bot binary are structurally unable to address cyborg propaganda
Core premise used to justify the need for new regulatory responses

invented entities (1)

cyborg propaganda no independent evidence
purpose: To name and analyze a hybrid human-verified plus AI-automated influence architecture
Newly coined conceptual category introduced to fill a perceived gap in existing theory

pith-pipeline@v0.9.0 · 5643 in / 1303 out tokens · 22292 ms · 2026-05-15T22:17:18.021139+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

frameworks built on the human/bot binary (including the EU AI Act and Section 230) are structurally unable to address
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

shifts political discourse from a contest of ideas to a battle of algorithmic campaigns

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.