SEC536: Adversarial AI - Penetration Testing AI Systems


Experience SANS training through course previews.
Learn MoreLet us help.
Contact usBecome a member for instant access to our free resources.
Sign UpWe're here to help.
Contact UsThe human collective, armed with critical thinking and shared ethical standards, remains the most powerful tool in the OSINT arsenal.

The SANS OSINT Summit and Training 2026, held in Amsterdam, and virtually, brought together the world's leading practitioners to address the "third generation" of open-source intelligence (OSINT), an era increasingly dominated by Artificial Intelligence (AI) and complex information operations. As the industry matures, the summit highlighted a critical tension: we have more data and better tools than ever, but the human element of rigorous analysis is being eroded by several factors.
Keynote speakers emphasized that OSINT is currently at a crossroads. Deniz Mykola Dirisu noted that the field has transitioned from traditional government archives (1st Gen) and digital proliferation (2nd Gen) into a 3rd generation characterized by AI-driven capacity and state-sponsored poisoning of information.
Nico Dekens challenged the community to move past what he calls "buttonology," the tendency to rely on automated tools without understanding their underlying logic. He argued that OSINT is a thinking game, not a tool race, and warned that a reliance on AI-generated summaries can lead to "cognitive decline" and the erosion of traditional tradecraft.
The summit showcased the cutting edge of automation while providing stark warnings about its risks:
The landscape of OSINT is shifting to counter increasingly sophisticated adversaries:
In one of the summit's most down-to-earth sessions, Lisette Abercrombie advocated for celebrating mistakes as essential learning moments. She shared critical lessons on OPSEC and methodology, including:
The overarching theme of the 2026 summit was the protection of cognitive security. As we face restricted APIs and the potential "death of the open web," practitioners must prioritize provenance over plausibility. The consensus was clear: while we should embrace AI as an accelerant, we must never outsource our judgment to machines.
The human collective, armed with critical thinking and shared ethical standards, remains the most powerful tool in the OSINT arsenal.


Launched in 1989 as a cooperative for information security thought leadership, it is SANS’ ongoing mission to empower cybersecurity professionals with the practical skills and knowledge they need to make our world a safer place.
Read more about SANS Institute