Full Report
This essay was written with Nathan E. Sanders, and originally appeared in The Guardian. There are plenty of signs that AI will make all of our experiences of the US midterm elections worse. Voters have anxiety about AI’s impacts on the country. Politicos are using AI deepfakes to spread lies. The White House is posting slopaganda. Meanwhile, candidates are missing a real opportunity to use AI to make campaigning better. The technology can help candidates listen more deeply to voters’ concerns, engage constituents more inclusively, and formulate policy platforms that are more responsive to our input. There are vanishingly few examples of this in ...
Analysis Summary
# Morning News Roll-up September 17, 2026
## Overview
Current reporting highlights a shift in the electoral threat landscape. While AI-driven disinformation, deepfakes, and "slopaganda" continue to threaten the integrity of the US midterm elections, emerging "pro-democracy" AI tools are being deployed by international actors to facilitate transparent constituent engagement and large-scale policy deliberation.
## Top Stories
### AI Deployment in Political Campaigns: Threats and Opportunities
- Summary: The report identifies a dichotomy in political AI usage. Negative applications include the proliferation of AI deepfakes to spread misinformation and the use of "slopaganda" (low-quality AI-generated propaganda) by official entities. Conversely, international groups like Japan's Team Mirai are utilizing AI for "broad listening," using chatbots to conduct 16,000+ deep-dive interviews with voters to shape legislative policy.
- Source: hxxps://www[.]schneier[.]com/blog/archives/2026/09/how-candidates-could-use-ai-for-good[.]html
### Emerging Civic Technology and Open-Source Policy Tools
- Summary: New platforms are emerging to facilitate "many-to-many" digital deliberation. CrownShy’s "Comhairle" tool and Stanford’s "deliberation[.]io" are highlighted as open-source frameworks designed to synthesize diverse viewpoints into actionable policy, providing a structured alternative to the "one-to-many" broadcasting methods typically used in digital influence operations.
- Source: hxxps://www[.]theguardian[.]com/commentisfree/2026/aug/31/ai-politics-voters
### The Rise of "Slopaganda" and Deepfake Disinformation
- Summary: Threat actors and political organizations are increasingly leveraging Generative AI to inundate voters with automated messaging and deceptive media. The White House and various political campaigns are noted for using AI-generated imagery and deepfakes, contributing to high levels of voter anxiety regarding the authenticity of election-related content.
- Source: hxxps://www[.]theguardian[.]com/us-news/2026/jan/29/the-slopaganda-era-10-ai-images-posted-by-the-white-house-and-what-they-teach-us
---
# Main Topic
Analysis of AI-driven influence operations in US midterm elections, contrasting malicious "slopaganda" and deepfakes with emerging pro-democracy deliberative technologies.
## Key Points
- **Weaponized Disinformation:** AI deepfakes are being actively used to spread lies and manipulate voter perception.
- **Slopaganda:** The rise of low-effort, AI-generated political propaganda (slopaganda) used by official and unofficial political entities.
- **Broad Listening:** A novel application of AI (pioneered by Team Mirai) that uses LLM-based interviewers to collect and synthesize complex voter sentiment at scale.
- **Shift in TTPs:** Transitioning from "one-to-many" broadcast-style manipulation to "one-to-one" or "many-to-many" engagement models.
## Threat Actors
- **Political Operatives:** Using deepfakes for character assassination and spreading false narratives.
- **State-Linked Entities:** Accused of utilizing "slopaganda" to flood information environments.
- **Civic Hackers (Neutral/Pro-Democracy):** Groups like Taiwan’s "gov zero" and Japan’s "Team Mirai" whose tools, while intended for transparency, represent a significant shift in how political influence is brokered.
## TTPs
- **AI Deepfakes:** Generation of synthetic audio/visual content to impersonate candidates.
- **Slopaganda:** Mass-generation of AI images and text to saturate social media feeds.
- **Automated Interviewing:** Utilizing LLMs to engage constituents in lengthy, structured data-gathering conversations.
- **Consensus Synthesis:** Using AI to analyze thousands of diverse perspectives to find common policy ground (e.g., Polis, Comhairle).
## Affected Systems
- **Electoral Integrity:** The US midterm election cycle and voter trust.
- **Social Media Platforms:** Primary vectors for AI-generated deepfakes and slopaganda (e.g., X/Twitter).
- **Public Discourse Infrastructure:** Town halls and legislative hearings now integrating AI-synthesized constituent data.
## Mitigations
- **Open-Source Transparency:** Utilizing open-source AI tools (e.g., CrownShy) to ensure the logic of constituent engagement is auditable.
- **Legislative Action:** Proposals for holding AI companies responsible for model-generated harms and taxing AI revenues.
- **Authenticity Verification:** Increasing voter literacy to identify synthetic media and "slopaganda."
- **Direct Engagement:** Moving away from broadcast ads toward verifiable, two-way AI-assisted deliberation to reduce the impact of top-down disinformation.
## Conclusion
The threat landscape for the US midterms is dominated by the ease of generating deceptive content. However, the emergence of "broad listening" tools suggests that the same technology used for disinformation can be repurposed for defensive, transparent democratic engagement. Analysts should monitor the adoption of open-source deliberative tools as a potential counter-measure to centralized AI-driven propaganda.