Visual Veracity — True (VIS) Lies: Analyzing How Generative AI Recognizes Intentionality, Rhetoric, and Misleadingness in Visualization Lies

📄 Research Insight

Problem Statement

This research addresses the challenge of detecting and understanding misleading visualizations in information dissemination, particularly in the context of COVID-19 communication.

Core Innovation

The study introduces a novel taxonomy of authorial intents that aids in the interpretation of intentionality behind misleading visuals.

In Plain English

This research explores how advanced AI can spot misleading graphics and understand why they might be created. It uses a collection of tweets and examples from a visualization event to test various AI models. The goal is to improve how we identify and interpret deceptive visual information.

Real-World Applications

• Enhanced fact-checking tools
• Educational resources for data literacy
• AI-driven content moderation systems

💡 Product Idea

Visual Veracity

Unmasking the truth behind data visualizations.

Visual Veracity is an AI-powered tool that analyzes visual content to detect misleading elements and assess their intent. By leveraging cutting-edge LLMs, it helps users understand and verify the integrity of the data presented in visualizations.

🚀 Execution Plan (MVP)

week 1 2: Develop a prototype that analyzes a small dataset of visualizations for misleading content.

week 3 4: Integrate user feedback and expand the dataset to include various domains beyond COVID-19.

week 5 8: Finalize features and prepare for a public launch, focusing on user interface and accessibility.

📊 Business Model

Target Market

• Primary: Journalists and fact-checkers
• Secondary: Educators and students in data science
• Market_size: Estimated TAM of $1 billion in data verification and education tools.

Revenue Model

• Primary: Subscription-based access for individual users and organizations
• Secondary: Partnerships with educational institutions for tailored solutions
• Pricing hint: Tiered pricing starting from $29/month for individuals.

🌍 Future Impact (5–10 Years)

In 5-10 years, this technology could significantly enhance public understanding of data, reduce misinformation, and promote transparency in data-driven decision-making.

📎 Original Paper:

True (VIS) Lies: Analyzing How Generative AI Recognizes Intentionality, Rhetoric, and Misleadingness in Visualization Lies
Authors: Graziano Blasilli, Marco Angelini
Categories: cs.HC, cs.CL, cs.CV
Published: April 1, 2026