AI-driven threats force evolution in cybersecurity strategies

Organizations are facing a paradigm shift in cybersecurity as AI-driven threats, such as deepfakes and automated social engineering, bypass traditional security silos. An EY survey highlights that many companies are still using outdated mindsets to combat modern risks, with only 12% of organizations feeling prepared to detect targeted physical attacks.

As AI agents and automated processes become more prevalent, a new market for agent-specific security is emerging. Investors have poured $435 million into AI agent security startups over five months, including a $50 million seed round for AIR Security. Experts suggest that AI agents must be treated as privileged identities with strict governance, least-privilege access, and continuous monitoring.

Technological solutions are evolving to address these gaps. Hummingbirds AI and IGEL have introduced Guacamole™ to provide continuous human verification during active sessions to prevent unauthorized access after login. Meanwhile, in the Asia-Pacific region, 51% of organizations are prioritizing digital sovereignty and tighter data control to mitigate security and compliance risks associated with rapid AI adoption.