Aletheia is an open source intelligence platform that unifies the open web and social media into a single, searchable view, helping investigators, governments, and newsrooms uncover meaning, mood, and motive across 30+ languages. It combines semantic search, narrative mapping, stance detection, and event extraction to turn raw data into citable facts. The platform supports live and archived sources including Telegram, RSS, YouTube, and X/Twitter, with custom ingestion adapters built on demand.
Funding
Funding not disclosed
Founders
Product
Problem
Investigators, journalists, and government analysts face an overwhelming volume of open-source and social media data spread across multiple platforms and languages. Manually tracking narratives, claims, and events across dozens of tabs is time-consuming, error-prone, and often misses critical connections hidden in non-native languages or unstructured media.
Solution
Aletheia provides a unified open source intelligence platform that aggregates the open web and social media into a single, searchable workspace. The platform supports live and archived collection from sources like Telegram, RSS, YouTube, and X/Twitter, with custom ingestion adapters available for unlisted sources. Users can search by meaning rather than exact keywords, enabling cross-lingual discovery of content with no keyword overlap. Aletheia automatically maps narratives, tracks stances on specific claims, and extracts typed, dated, and confidence-scored events, all presented through an interactive visual interface. The platform also offers an open API for custom analysis and integrates automatic translation while preserving original language for verification.
Target Audience
Primary users are government intelligence analysts, investigative journalists, newsrooms, and research organizations that need to monitor, analyze, and cite open-source information at scale.
Features
- Semantic search across 30+ languages that matches content by meaning, not exact words, enabling cross-lingual discovery without keyword overlap
- Image-based search to match photos across video and stills
- Narrative mapping that clusters related articles, identifies keywords, people, places, emotional tone, and volume trends over time
- Stance detection that classifies mentions of tracked claims as support, oppose, mixed, or unclear, with evidence links for each judgement
- Event extraction that automatically identifies typed, dated, and confidence-scored events from collected articles
- Live and archived source ingestion for Telegram, RSS, Web, X/Twitter, YouTube, Facebook, and TikTok, with custom adapters built on demand
- Automatic translation with original language preserved for verification
- Open API for custom analysis and integration into existing workflows