🔒 About the Hackathon

As AI systems advance toward general intelligence capabilities, the security implications become increasingly critical. This hackathon, organized provides a practical forum for participants to develop and test novel approaches to securing AI systems and using AI to improve cybersecurity measures.

💡 Challenge Tracks

1. Classical Cybersecurity for AI: Work on developing robust protection mechanisms for AI infrastructure and model weights. Focus on creating approaches that safeguard training environments, prevent model theft, and secure the entire AI development pipeline.

2. AI for Cybersecurity: Leverage AI capabilities to enhance traditional cybersecurity domains. Design systems for automated vulnerability discovery, intrusion detection, threat hunting, and intelligent patching mechanisms that outperform conventional approaches.

3. Model Security: Design and implement innovative guardrails, safeguards, and protection mechanisms for AI models. Focus areas include preventing PII leakage, defending against adversarial inputs, ensuring behavioral safety for agentic models, and creating robust monitoring systems.

👥 Who Should Participate

We welcome participants from diverse backgrounds, including:

• Cybersecurity professionals

• AI researchers and engineers

• Red team specialists

• Privacy experts

• Systems security engineers

• Policy researchers interested in AI security governance

No prior experience with AI security specifically is required, though familiarity with either machine learning, cybersecurity, or related fields is helpful.