🧪 This platform is in early beta. Features may change and you might encounter bugs. We appreciate your patience!
Because fact-checking tells you what is wrong. We want to understand how and why reasoning fails.
Every day, billions of arguments are made — in boardrooms and parliaments, on social media and in news broadcasts, in advertising and in private conversations. Some are sound. Most are not. And the gap between the two is where manipulation lives.
Fact-checkers tell us whether a claim is true or false. That's important. But it doesn't explain why a flawed argument felt convincing. It doesn't reveal the structure of the deception. It doesn't name the technique being used.
Consider this statement:
"Millions of parents are concerned about vaccines. If that many people worry, there must be something wrong. Scientists who disagree are obviously funded by pharma."
A fact-checker would label this "misleading." But inside these three sentences hide at least four distinct reasoning flaws: an argumentum ad populum, a hasty generalization, an ad hominem, and a genetic fallacy. Each exploits a different cognitive weakness. Each requires a different defense.
TellDear exists because naming the error is the first step to disarming it.
We don't judge whether a claim is true. We examine whether the reasoning behind it is valid. A true conclusion can still be supported by terrible logic. A false conclusion can be wrapped in seemingly perfect argumentation. Both deserve scrutiny.
Campaign speeches, parliamentary debates, policy proposals — dissect the rhetoric behind political positions. Identify when leaders use emotional manipulation instead of evidence.
Board presentations, strategy memos, investor pitches — reveal when business arguments rest on faulty logic, cherry-picked data, or false dilemmas.
Editorials, opinion pieces, news framing — go beyond what is said to analyze how it is said. Detect loaded language, false balance, and narrative manipulation.
Teach students not just what to think, but how to evaluate thinking. A searchable taxonomy of 535+ reasoning patterns as a teaching resource.
Viral posts, threads, comment wars — identify the manipulation techniques that make misinformation spread. Understand why certain arguments feel compelling despite being hollow.
Everyday conversations, family arguments, negotiation — recognize when you or others fall into cognitive traps. Better reasoning starts with self-awareness.
TellDear is part of a growing movement that goes beyond fact-checking into structural analysis of arguments. Here's who else is working in this space:
AI system that competes in structured debates against humans — using claim detection, evidence mining, and argument quality assessment as standalone APIs.
Annual academic workshop since 2014, bundling research on automatic argument recognition, claim extraction, and argument quality assessment (ACL/EMNLP).
Benchmark competition classifying 14 propaganda techniques — treating propaganda not as a truth problem but as a technique problem. 44 teams participated.
Stephen Toulmin decomposed arguments into Claim, Grounds, Warrant, Backing, Qualifier, and Rebuttal — the intellectual foundation of modern argument mining and AI training.
Structured debate platform visualizing arguments as pro/contra trees. Over 1M users, used at Harvard and Princeton.
Open-source Markdown-like syntax for argument reconstruction and graph visualization.
Open-source collaborative argument mapping software for academic use in philosophy.
News aggregator showing coverage from left, right, and center — plus "blindspot" analysis of what each side ignores.
Same news from different political directions side by side. Shows how framing changes perception of identical facts.
Interactive chart mapping news sources by factuality and political bias.
Combines AI-powered NLP with human fact-checkers to monitor 1M+ domains. Classifies disinformation campaigns and narrative structures.
Free e-learning platform teaching bias recognition, conspiracy pattern detection, and rhetorical techniques in media.
Policy research analyzing the structural erosion of factual foundations in public discourse — opinion-fact blurring, trust decline, volume shift to opinion.
Popular reference card of 24 classical reasoning errors. Mass-market entry point for logical fallacy awareness.
Most existing tools focus on one slice: logical fallacies or propaganda or media bias. TellDear integrates all of these into a single, machine-readable taxonomy with 535 aspects across 6 dimensions.
The best defense against bad reasoning is the ability to recognize it. Try the analyzer, explore the taxonomy, understand the patterns.