Webinar Description
Key Takeaways
- Webinar exploring how offensive security teams can develop capabilities for assessing AI systems
- Features firsthand insights from an OSAI Early Access Program participant at Splunk, a Cisco company
- Addresses the challenge of integrating AI security testing into established penetration testing workflows
- Designed for security leaders, offensive security managers and penetration testers
- Focuses on establishing consistent methodologies for AI system assessment across teams
Introduction
Inside OSAI EAP: Preparing Your Team for AI Security is a webinar hosted by OffSec that examines how offensive security teams can adapt their practices to address the distinct challenges posed by AI systems. The session is aimed at security leaders, penetration testers and red team professionals who are evaluating how to incorporate AI security assessment into their existing operations. As organisations increasingly deploy machine learning models and AI-driven applications across critical business functions, the attack surface has expanded in ways that traditional penetration testing methodologies were not designed to address. This webinar arrives at a time when security teams are grappling with how to evaluate AI systems for vulnerabilities that differ fundamentally from conventional software flaws.
About This Event
The webinar centres on a live conversation with a participant from the OSAI Early Access Program, OffSec’s training initiative focused on AI security. The featured speaker represents Splunk, a Cisco company, and shares practical experiences from integrating AI security training into an operational penetration testing team. Rather than presenting theoretical frameworks, the session draws on real-world implementation challenges and the lessons that emerged from applying new assessment techniques within an enterprise security function.
The virtual format allows for interactive discussion, enabling attendees to engage directly with insights from an organisation that has already navigated the early stages of building AI security capabilities.
Integrating AI Security into Offensive Operations
A central theme of the webinar is the practical challenge of weaving AI security assessment into existing penetration testing workflows. Offensive security teams have well-established processes for evaluating traditional applications, networks and infrastructure. AI systems, however, introduce novel attack vectors including prompt injection, model manipulation, training data poisoning and adversarial inputs that can cause models to behave unpredictably.
The session addresses how teams can develop a shared methodology for testing AI systems, ensuring that assessments are consistent and repeatable rather than ad hoc. This methodological consistency becomes particularly important as AI deployments scale across an organisation and multiple team members need to conduct assessments using comparable approaches.
Bridging the AI Security Knowledge Gap
Many offensive security professionals possess deep expertise in network penetration, application security and infrastructure testing, yet AI security represents a relatively new domain requiring different technical knowledge. Understanding how machine learning models function, where they are vulnerable and how adversaries might exploit them demands familiarity with concepts that sit outside traditional security training.
The webinar explores how the OSAI program helps teams bridge this knowledge gap, providing structured training that enables penetration testers to extend their capabilities into AI assessment. For security leaders, the session offers practical guidance on preparing teams for this transition without disrupting existing operations.
Who Should Attend
The webinar is designed for security leaders responsible for offensive security functions, managers overseeing penetration testing and red team operations, and individual practitioners seeking to understand how AI security fits within their professional development. Organisations currently participating in the OSAI program or evaluating AI security training options will find the firsthand implementation insights particularly relevant. The content assumes familiarity with offensive security concepts and is oriented toward professionals working in enterprise environments where AI adoption is accelerating.
The Growing Importance of AI Security Assessment
As AI systems move from experimental deployments into production environments handling sensitive data and critical decisions, the security implications have become impossible to ignore. Offensive security teams are increasingly expected to assess these systems alongside traditional infrastructure, yet many lack the specialised knowledge required. The webinar positions AI security as an essential component of modern offensive security practice rather than a separate discipline, reflecting the reality that AI systems are now embedded throughout enterprise technology stacks and require the same rigorous security evaluation as any other business-critical application.