Malicious AI Agents Disrupt APAC Enterprises: New Threats

Malicious AI agents are severely disrupting enterprises in the Asia-Pacific region. Attackers are using these agents to steal credentials, spread malware, and compromise systems while remaining undetected. Here's the thing: autonomous AI systems are making APAC companies more vulnerable. Attackers are bypassing governance and triggering complex workflows. A study found only nine percent of companies are prepared for these AI-driven threats. What's interesting is how attackers are shifting their focus. Instead of targeting human logins, they're exploiting AI agents that already have legitimate access to systems, data, and workflows. These agents are designed to operate autonomously, which introduces new security risks. For example, a compromised AI agent with broad administrative access could be hijacked to delete a firm's database or expose sensitive user records. This happens because traditional identity models authenticate agents once and issue a static token, assuming nothing will change. Attackers exploit this lack of real-time control. The bottom line is that these compromised agents can execute actions across multiple systems much faster than human attackers. This means firms need to rethink their security strategies as they deploy agentic AI.