| Mobile Web

Search results for prompt injection

AI & Enterprise

First-half cyber intrusion reports up 20 percent in South Korea as DDoS and ransomware surge

Reports of cyber intrusion incidents in South Korea rose about 20 percent in the first half of 2026 from a year earlier, a government report showed. The Korea Internet & Security Agency received 1,236 cases, up 19.5 percent, as DDoS attacks and ransomware increased sharply. The ministry and KISA also flagged threats linked to the spread of generative AI, open-source supply-chain attacks, and data leaks exploiting APIs and stolen accounts.

AI & Enterprise

KAIST selected for Microsoft\'s global AI safety research program

KAIST has been selected for Microsoft\'s global AI safety research program, the only South Korean university chosen. KAIST said computer science professor Samuel Sohn (손수엘) was picked for research support under the \'External Red Teaming Alliance (EXTRA)\' run by the Microsoft AI Red Team. Launched this year, the program tests AI systems for security flaws, misuse risks and alignment issues. KAIST will receive $25,000 in funding for AI safety, security and responsible AI research.

AI & Enterprise

Harness launches \'Agent DLC\' to deploy AI agents with familiar development processes and tools

Harness has launched \'Agent DLC\' to help companies run AI agents within existing software development and deployment systems. The product integrates agent-specific evaluation, deployment governance and security controls to reduce issues when moving agents from concept to production. It lets teams build, test, deploy and operate agents in existing CI/CD pipelines. Harness said only 8 percent of organizations have deployed agentic AI in production, citing unpredictability.