(C01) (INCYBER) Réussir sa transition vers l’Agentic SOC: le guide pratique par Google et I-TRACING

Value of the Information & Strength of the Argument

The video provides valuable insights into the practical application of AI in SOC operations, based on real-world experience from I-TRACING and Google. The argumentation is solid, with clear reasoning for why traditional SOCs are failing and how AI agents can address these issues. The speakers present a structured approach, from identifying the problem to implementing and measuring the solution. They emphasize the importance of data quality, context, and governance, which are critical for successful AI adoption. The live demo of a BEC investigation adds credibility and demonstrates the feasibility of the proposed approach. However, the presentation is somewhat promotional, and the technical depth is limited, with some concepts only briefly explained.

Scientific Rigor, Source Quality, Title Accuracy

The presentation is based on the speakers’ professional experience and their work with Google SecOps and I-TRACING. While they mention specific tools and technologies, they do not cite external sources or academic references. The title accurately reflects the content, as the video is indeed a practical guide to transitioning to an Agentic SOC. The information is presented in a clear and structured manner, but the lack of citations and the promotional tone reduce the overall scientific rigor. The video includes a brief interactive segment with the audience, but no comments are provided for analysis.

The title accurately reflects the content: a practical guide to transitioning to an Agentic SOC, presented by Google and I-TRACING.

7/10

The talk is based on practical experience from I-TRACING and Google, with concrete examples and a live demo. However, it is largely promotional and lacks detailed technical depth or peer-reviewed sources.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 02:00Introduction: Michel Schbert and Clément from I-TRACING, and Jonathan George from Google, introduce the topic of transitioning to an Agentic SOC.
• 05:00Discussion on the limitations of traditional SOCs: alert fatigue, complexity, and the need for AI.
• 10:00Jonathan George explains what an AI agent is and introduces the concept of MCP (Model Context Protocol).
• 15:00Overview of Google SecOps platform and its AI-driven features, including the alert investigation agent.
• 20:00Discussion on the importance of data quality and context for AI agents, and the need for customization.
• 25:00I-TRACING presents their custom agentic layer, including MCP integration and runbooks.
• 30:00Live demo of a BEC attack investigation using Google SecOps and custom AI agents.
• 35:00Detailed walkthrough of the BEC detection and investigation process, including triage, CTI enrichment, and email analysis.
• 40:00Discussion on the shift from deterministic playbooks to adaptive reasoning, and the importance of explicability.
• 45:00Conclusion: key takeaways and next steps for implementing an Agentic SOC.

• Forum INCYBER Europe — Mentioned as the event where the presentation took place.
• INCYBER Europe LinkedIn — Mentioned as a way to stay connected with the forum.

• Google SecOps — The platform used in the demo, consistent with the presentation.