While online communities provide continuous engagement, professional conferences and events offer concentrated bursts of knowledge, networking, and insight into the industry’s trajectory. For an AI red teamer, these gatherings are not just educational opportunities; they are intelligence-gathering missions. You can observe emerging attack vectors, gauge defensive priorities, and connect with the researchers and practitioners shaping the field.
The Landscape of AI Security Events
The events relevant to AI red teaming do not all fit under one banner. They span a spectrum from highly theoretical academic gatherings to hands-on, practical hacking conferences. Understanding where an event sits on this spectrum helps you align your objectives with the right venue. The most valuable insights often come from the intersections of these domains.
Figure 23.5.3-1: A conceptual map of relevant event types, highlighting the central role of specialized venues like the AI Village that bridge academic, corporate, and practical hacking communities.
Key Event Categories and What to Expect
Navigating this landscape requires a strategy. Each type of conference offers a different kind of value. Your goal should be to sample from each category over time to build a comprehensive understanding of AI security.
1. Core Cybersecurity Conferences
These are the large, established events in the information security world. While their focus is broad, they increasingly feature dedicated tracks, villages, and talks on AI/ML security. This is where you’ll see how AI vulnerabilities are being integrated into the wider threat landscape.
- Value Proposition: Broad industry context, practical tool releases, and networking with traditional security professionals who are now tackling AI challenges.
- What to Look For: Dedicated “villages” (like the AI Village at DEF CON), workshops on ML tool security, and talks on real-world exploits against production AI systems.
2. Dedicated AI/ML Security Workshops and Summits
These smaller, more focused events are where deep, specialized work is presented. They are often co-located with larger academic or security conferences. Attending these is essential for staying at the absolute cutting edge of research.
- Value Proposition: Direct access to top researchers, deep dives into specific attack modalities (e.g., membership inference, model inversion), and early previews of future threats.
- What to Look For: Events like the AISec workshop or specialized summits focused on topics like generative AI safety or LLM security.
3. Academic Machine Learning Conferences
Events like NeurIPS, ICML, and ICLR are the primary venues for foundational ML research. While 95% of the content is not directly about security, the remaining 5% is where novel vulnerabilities and theoretical attack frameworks are first published. Monitoring these events is like watching the forge where future weapons are made.
- Value Proposition: Understanding the fundamental principles behind new model architectures and their inherent weaknesses. This is long-term intelligence gathering.
- What to Look For: Sessions on “Robustness,” “Privacy,” “Adversarial Examples,” and “Trustworthy AI.”
Reference Table of Key Conferences and Events
The following table provides a non-exhaustive list of significant events for AI security professionals. Note that the field is dynamic, and new events emerge regularly.
| Conference / Event | Primary Focus | Relevance to AI Red Teaming | Typical Audience |
|---|---|---|---|
| DEF CON (AI Village) | Practical Hacking & Security Culture | The epicenter of public AI red teaming. Features hands-on exploits, live LLM hacking competitions (e.g., GRT), tool releases, and community building. Essential for practical skills. | Hackers, Red Teamers, Researchers, Hobbyists |
| Black Hat USA / EU / Asia | Corporate Cybersecurity | Cutting-edge briefings on enterprise threats, vendor solutions, and defensive strategies. More formal than DEF CON, with excellent (but expensive) training sessions on AI security. | Security Professionals, C-Suite, Vendors, Researchers |
| AISec Workshop | Academic AI Security Research | A premier academic venue for formal research on attacks and defenses. Often co-located with a major security conference like ACM CCS. Ideal for understanding theoretical foundations. | Academics, PhD Students, Corporate Researchers |
| NeurIPS / ICML / ICLR | Core Machine Learning Research | Discover foundational model weaknesses and theoretical attack vectors before they are widely weaponized. Requires actively seeking out security and robustness tracks. | ML Researchers, Data Scientists, Academics |
| RSA Conference | GRC & The Business of Security | High-level strategy, policy, and the business of AI security. Useful for understanding AI risk management, compliance (e.g., EU AI Act), and organizational priorities. | CISOs, Managers, Policy Makers, Vendors |
| USENIX Security Symposium | Systems Security Research | Bridges theory and practice. Often features high-quality, peer-reviewed papers on practical attacks against ML systems, including software and hardware vectors. | Academics, Applied Researchers, Security Engineers |
Maximizing Your Conference Experience
Simply attending is not enough. To extract maximum value, you need a plan:
- Prioritize Workshops and Villages: The most valuable interactions often happen in smaller, more focused settings. A two-hour workshop can be more impactful than a full day of 30-minute talks.
- Engage in CTFs and Competitions: Participating in events like the Generative Red Teaming (GRT) challenge at the AI Village is one of the fastest ways to benchmark and improve your practical skills.
- Network with a Purpose: Don’t just collect business cards. Identify the researchers or practitioners whose work you admire and seek them out. Ask specific, insightful questions about their work. The “hallway track” is often the most valuable part of any conference.
- Review Proceedings Post-Event: You can’t attend every talk. After the event, download the slide decks and papers for sessions you missed. This is particularly important for academic conferences where the papers contain the full details.
By strategically selecting and engaging with these events, you can ensure your skills and knowledge remain at the forefront of the rapidly evolving field of AI security.