Skip to content


At the UDVL, we’re committed to advancing OSINT methods and best practices. Our in-house methods development team incubates new and emerging approaches to open-source investigation. This team conceptualizes, develops, and refines tools and processes for use in UDVL investigations and beyond. Promising methods and tools are shared for peer review, and mature tools and methods are made available to the OSINT community. We welcome collaboration and are expanding our membership this fall. The team includes experienced OSINT analysts, machine learning and software experts, and applied AI researchers.


ML based tools show promise in OSINT applications. They can sift through vast tranches of data – working around the clock without tiring. They are particularly useful in text applications and where projects demand visual analysis at scale. We are developing scalable ML tools to help our analysts go farther with the time available to them.


Damage assessment and damage evidence gathering are vital to post-conflict judicial processes. Accurate damage assessment is useful beyond the courtroom in informing reconstruction efforts and resource allocation. Because of this, the ability to perform damage assessment prior to the cessation of conflict in a locality gives responsible parties a head start. While conflict often prevents contemporaneous physical damage assessment of sites, remote assessment can be conducted regardless of human security concerns on the ground. In some cases, remote assessment can provide accurate and valuable insights. We’re working to establish methods and develop software tools to improve and standardize remote damage assessment.


To be useful, OSINT analysis must be rigorous and transparent. This goes double where information derived through OSINT is used in judicial proceedings or recovery and reconstruction planning. The standardization of methods, metrics, and vocabulary contribute to peer-to-peer transparency and reenforce the legitimacy of the field. Our team is working to create open methods standards intended to improve repeatability of results and to increase the transparency of our investigations.