How to Find WhatsApp and Telegram groups in the Open
Three layers of collection for narrative intelligence, extremism monitoring, and any investigation that needs an analyst inside the room.
Welcome back.
This week we walk through the techniques we use to find public WhatsApp and Telegram groups in the open, for narrative intelligence work, extremist monitoring, and any investigation where the analyst has to be inside the room before any content is visible.
Plus, the latest tradecraft tips and industry news from last week.
Missed last week?
Before we get into it, here is last week's tradecraft and industry news roundup:
OSINT Tradecraft Tips & Industry News
Everything you need to know from the past seven days.
GIJN’s latest guide walks you through QGIS, from the basics to advanced data analysis, with practical applications drawn from investigative work on crime hot spots, environmental destruction, demographic patterns, and war crimes | read here
Matthieu F. has developed a French-language tool to document and compare influence operations analysed from open sources, consisting of two standalone HTML tools usable offline: a structured input form based on the Diamond Model, DISARM Framework, and DIMA, and a viewer to visualise multiple operations simultaneously on an interactive timeline | read LinkedIn post | access GitHub
New updates released for the free browser-based tool Osintracker | explore here | test here
Google Labs has introduced three new exploratory tools for AI-powered scientific discovery:
Literature Insights (built with NotebookLM) for literature search and artifact generation
Hypothesis Generation (built with Co-Scientist) for multi-agent research ideation
Computational Discovery (built with AlphaEvolve and ERA) for generating and scoring code variations against optimisation metrics | explore here
TechCrunch reports that Google has unveiled an AI-powered overhaul of Search at its I/O conference, with a reimagined search box that drops users into interactive AI experiences, deploys “information agents” to gather data on their behalf, and lets users build personalised mini apps directly in Search. TechCrunch has since clarified that links are not entirely gone, following misinterpretations spreading online | read here
Geoff Khan argues for the Cipher Brief that the intelligence cycle cannot keep pace with the current environment, and that tradecraft, not technology, is now the limiting factor | read here
Theresia Tanzil breaks down what makes web scraping in Asia distinct, from super-app login walls and behavioural CAPTCHAs to sophisticated bot mitigation and a country-by-country legal patchwork, and what it takes to operate there at scale | read here
Mirko Lorenz argues for DW Innovation that the AI-mediated content economy is breaking the chain of provenance, attribution, and trust that holds professional content together | explore here
A new three-year research project, PADSE, brings together Fraunhofer IDMT, Bauhaus-Universität Weimar, and Deutsche Welle to develop a detection system for manipulated and partially synthetic audio content | read more
Tara Calishain shares three free, ad-free search tools she built as alternatives to Google’s AI-driven search: MiniGladys, Wikipedia Seismograph, and Wiki-Guided Google Search (I & II), each using Wikipedia to support link-based discovery and topical exploration. | view here
Finding public chat groups that do not want to be found
Chat platforms are difficult to monitor. Unlike X, where posts sit in the open and can be searched and visualised at scale, WhatsApp and Telegram groups require an analyst to be inside the room before any content is visible.
For groups that are genuinely private, the kind extremists set up among trusted contacts and circulate by word of mouth, there is rarely a way in from open sources. Those rooms are not the subject of this week’s post.
What we are looking at are groups that are meant to be public. Communities, movements, and networks that want members, and therefore have to share their invite links somewhere a stranger can find them. The link has to surface. The question is where, and how to find it. We use these techniques regularly for narrative intelligence work, for monitoring extremist ecosystems, and for any investigation that requires us to be inside a WhatsApp or Telegram room before we can see what is being said. The methodology transfers across use cases. The tools change. The logic of where a public invite link ends up does not.
