{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/8785a58ff0b74b7d83a906695d1c4712\" frameborder=\"0\" width=\"1728\" height=\"1296\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1296,"width":1728,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1296,"thumbnail_width":1728,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/8785a58ff0b74b7d83a906695d1c4712-95953a5d3b014e73.gif","duration":228.701,"title":"TruthLens AI Checks Claims with Live Research","description":"This Loom explains how TruthLens AI verifies online claims using live web research. The presenter says TruthLens uses the U.com Research API to investigate a user-entered factual claim, then organizes results into supporting evidence, conflicting evidence, official confirmation, and a story timeline with an explainable confidence score. An example involving a rumor about Google acquiring something was assessed with 83% of evidence supporting it, but not 100%. Another example, “OpenAI is currently building its own smartphone,” was deemed misleading with an 88% misleading score, supported by supporting and official evidence. The system is connected through FASTAPI endpoints for live investigation."}