KISA and SANDS Lab host briefing session for the '2025 Cybersecurity AI Dataset Demonstration Project'

AI security firm SANDS Lab (KOSDAQ 411080, CEO Kim Ki-hong) announced it will hold the '2025 Cybersecurity AI Dataset Demonstration Participation Briefing' on the 18th at the Korea Internet & Security Agency (KISA) headquarters in Seoul.

Hosted by KISA and organized by SANDS Lab, the briefing is aimed at officials from domestic companies and institutions who want to adopt AI in security operations but are struggling to put it into practice, walking them directly through the on-site demonstration results and use cases of AI incident-response models.

At the briefing, SANDS Lab will introduce the demonstration at the core of the 'Cybersecurity AI Dataset Modernization and Utilization Enhancement' project and run live demonstrations for prospective users. Participating organizations can experience the AI incident-response models through four methods: API integration, a dedicated UI (KISA C-TAS), and dataset provisioning.

This year's AI dataset demonstration in particular focuses on showing how AI-based detection, analysis, and response processes can be applied to real security-operations environments — validating threat-response models such as determining whether EXE/PDF/ELF files are malicious, determining whether a domain is malicious, and identifying APT attack groups, all in practical settings.

In Shin-gyo, the SANDS Lab director overseeing the cybersecurity AI dataset project, said, "Real-time threat response isn't just a matter of technology development — it's a paradigm shift in the national security system. We hope this demonstration leads to the advanced AI dataset being actively used in practice, becoming a solid foundation for building out AI security response systems."

The briefing will run in two sessions, morning and afternoon, and is expected to draw security officials from a range of industries including small businesses and startups.

Over the past four years, KISA has built a cumulative 2.3 billion-record AI security dataset through its AI dataset project, and SANDS Lab has carried out the project for five consecutive years, winning a total of roughly 14.8 billion won in national contracts.