Cybersecurity Incident Response Drill for Internal AI Systems

The participating teams focus on handling simulated scenarios involving cyberattacks, defense measures, and cybersecurity protection for internal AI systems

The drill brought together over 70 officers and specialists from 10 participating teams. This new focus area centers on internal AI systems, a foundational technology that Hue City is planning to invest in and develop in the coming years.

The drill was conducted on-site under the theme Capture The Flag (CTF) and took place right at the conference hall. Participating teams worked through a series of simulated scenarios involving cyberattacks, defense measures, and the protection of cybersecurity and data security throughout the development, deployment, and operation of internal Large Language Model (LLM) system.

The simulated scenarios highlighted various information security threats, including malware embedded in open-source model supply chains, training data poisoning, model theft and reverse engineering, prompt manipulation, sensitive data leakage, identity theft, unauthorized use of AI accounts, and distributed denial-of-service (DDoS) attacks.

Based on these scenarios, participating teams were required to assess potential risks and vulnerabilities at each stage of system development, then propose and implement technical measures to ensure cybersecurity and data protection. At the same time, they also had to prepare incident reports and offer recommendations for remedial solutions.

The exercise provided specialized personnel with an opportunity to strengthen their ability to identify vulnerabilities and analyze potential risks at every stage of AI system development, from design and development to operation. It also helped them proactively develop appropriate technical solutions and defense strategies.