Study: Slanted framing of real images outpaces deepfakes in multimodal misinformation
IAugenstein · x · 2026-09-02
An analysis of 27k posts across 7 languages on X reveals the mechanisms of multimodal misinformation. It finds that slanted framing of real images is more prevalent than deepfakes or doctored images, while vaccine misinformation frequently borrows credibility via news screenshots. The paper introduces a unified taxonomy (MMMMM) and an automated annotation pipeline using Vision-Language Models (VLMs) to validate these insights.
More from Safety
- One diagram explains the difference between CSRF and XSS attacks — _jaydeepkarale · 2026-10-18
- Grady Booch slams Anthropic for tolerating up to 50 deaths and $1B damages per model — Grady_Booch · 2026-10-18
- Single Neuron Suffices to Bypass Safety Alignment in LLMs, Paper Finds — burny_tech · 2026-10-18
- Polymarket gives OpenAI just 8% odds of a full training pause by end of 2026 — Polymarket · 2026-10-18
- OpenAI engineer over alleged hidden recording device seeks bail, denies Saudi escape plan — Polymarket · 2026-10-18
- Free comic "The Collective" offers a strange retelling of the Hugging Face hack — Flugelwagen · 2026-10-18