"AI systems help US intelligence community with decision making"

Western governments, especially in the national security and intelligence sphere, are facing an asymmetrical threat as AI has become readily adopted by criminals and hostile governments. Check out the article I co-authored with Jay Heisler for Canadian International Council's Open Canada publication discussing the urgent need to find ways past the pervasive mistrust and risks of relying more extensively on AI for #OSINT investigations. https://lnkd.in/ecB2i5mk

Intelligence used to be about secrets. Now it’s about data - who holds it, who can process it, and who can move fastest. As AI transforms the field, OSINT isn’t just a supplement to classified work - it’s becoming the foundation. This week’s OSINT Newsletter highlights an essential read from Foreign Policy on why OSINT should be treated as a national asset. Read the full piece: https://lnkd.in/egVWCTKZ And read about more OSINT tools and techniques in Issue 83 of The OSINT Newsletter: https://lnkd.in/eYQcGGGD

📘 An insightful and timely read on the future of intelligence. Greg L. (CEO, Strider Technologies) and Dr Calder Walton (Harvard Kennedy School) make a powerful case for why open-source intelligence must be treated as a national asset. As the article argues, the intelligence race is no longer defined by who holds the most secrets, but by who can harness, analyse, and act on open data the fastest. AI is accelerating this shift, turning OSINT into a strategic capability that sits at the centre of modern intelligence and counterintelligence. For anyone working in security, defence, or data analysis, this is essential context on where the field is heading. Read the full piece below 👇

The intelligence landscape is undergoing its most profound shift since the Cold War. With the explosion of openly available data and the rise of AI systems capable of analyzing it at scale, intelligence is no longer defined by secrecy but by speed, accessibility, and insight. Open-source intelligence (OSINT) is emerging as the foundation of modern intelligence — not a supplement to classified collection, but its starting point. In an era where most of the world’s data exists in public digital environments, the central question has changed from “What can we uncover?” to “What’s already visible?” This transformation is reshaping who leads in intelligence. Increasingly, the most sophisticated capabilities in data analysis, threat detection, and counterintelligence come from the private sector, where innovation in AI and data science moves faster than government systems. Corporate and academic teams have already demonstrated they can detect state-linked operations — from cyber intrusions to full-scale invasions — ahead of national agencies, using only open and commercial data. The future of intelligence and counterintelligence will depend on who can combine open data, AI, and cross-sector collaboration to deliver actionable foresight faster than their rivals. The advantage now belongs not to those who guard secrets, but to those who can interpret the world in real time. #OSINT #AI #Intelligence #Counterintelligence #Geopolitics #NationalSecurity #InformationDominance #DataAnalysis https://lnkd.in/egVWCTKZ

AI deepfakes are reshaping wartime propaganda— and OSINT techniques must evolve with them. Over the past weeks, AI-generated videos of "Ukrainian soldiers" crying, surrendering to Russia, or complaining about mobilization have flooded social media. They look (kind of) convincing. They sound real. They're completely fake. In my latest Knowmad OSINT article (long overdue I know!), I break down how as part of their Cognitive Warfare, Russian PSYOPS campaigns are weaponizing AI video generators like Sora and Veo to undermine morale and manipulate Western audiences—and more importantly, how OSINT professionals can detect them. Without going into technical details or aiming for an all-comprehensive resource, in this article I cover: ▶️The telltale signs of deepfakes; ▶️Tools that can help; ▶️Why the old OSINT rule still holds true: online tools can help, but our strongest instruments remain our own mind and eyes. As adverserial Cognitive Warfare enters this new phase, teaching AI-centered media literacy and OSINT verification workflows to the public and specialized audiences has never been more critical. https://lnkd.in/ej7XRX_d

Our Head of Intelligence breaks down a critical evolution in modern warfare: Russia's deployment of increasingly sophisticated #AI-generated deepfakes targeting Ukrainian soldiers and Western audiences. These sophisticated fake videos—showing fabricated scenes of surrender, complaints, and demoralization—represent a dangerous new phase of Cognitive Warfare designed to erode morale and diminish international support for #Ukraine. As I. Vlad Sutea warns in his latest analysis, this marks the exacerbation of a trend that previously focused on "cheap fakes" and out-of-context forgeries, now moving towards increasingly "authentic" fabrications.

New research from Anthropic shows that even a handful of malicious samples, just 250 documents, can poison large language models of any size. A powerful reminder of why data security and model integrity matter at every scale 🔗 https://bit.ly/48KAFUT