Park Chan-am: AI Speeds Cyberattacks, Urges New Defenses

White hat hacker Park Chan-am is highlighting key security challenges in the AI era. He's the CEO of Steelion and a cybersecurity advisor for Korean government agencies. Here's the thing: Park warns that AI significantly speeds up cyberattacks. He says finding vulnerabilities used to take at least four weeks, but now it can take less than a day. This makes all internal software a critical target. What's interesting is his emphasis on securing internal software and the software supply chain. He stresses that once software is installed, it becomes part of the corporate system, not just a vendor issue. Park also points to authorization management as a major challenge, especially with Model Context Protocol-based AI agents. These agents can create new attack vectors if they have excessive privileges or encounter malicious documents. The bottom line: As AI advances, businesses need to rethink their security strategies, focusing on access control, supply chain management, and human verification to protect their systems.