The Problem
We live in an era of information overload. Every day, millions of articles compete for attention — and the incentives of modern media too often reward speed, outrage, and confirmation bias over accuracy and nuance.
It's increasingly hard to know what's actually true, what's missing from the story, and whether the framing you're reading reflects reality or a particular agenda. Most people don't have time to cross-reference every claim or read multiple sources for every article.
The Tools
PKL Media Lab builds AI-powered tools that give you a fast, independent second opinion on what you're reading — so you can consume news more confidently and critically.
How They Work
TruthPrism extracts 3–6 key factual claims from an article, then verifies each one in parallel using real-time Brave Search — an independent web index that provides fresh, unbiased results. Claude AI evaluates the evidence and assigns a verdict for each claim, then scores the article for both factual accuracy (Factual Score) and contextual fairness (Context Score).
News-Distiller uses Claude AI to read an article and produce a structured summary — an Executive Summary for quick reading or a Full Report when you want every detail. It also evaluates political lean, requiring direct evidence from the article text rather than inference.
Both tools stream results in real time so you see the analysis as it happens, rather than waiting for a single response.
The Technology
Both apps are built on Claude AI by Anthropic — one of the most capable and safety-focused large language models available. TruthPrism also uses Brave Search for real-time web verification, ensuring claim verdicts are grounded in current sources rather than AI training data alone.
AI analysis can make mistakes. Both tools are designed to assist critical reading, not replace it. Always verify important claims through multiple independent sources.
About the Builder
PKL Media Lab is the work of Paul Laikind, PhD — a retired biotech entrepreneur and executive. These apps were built through AI collaboration as a personal project driven by a belief that good tools for evaluating information should be accessible to everyone.