AI is transforming cybersecurity by enabling faster detection and response to threats, but it also accelerates cyberattacks like phishing and data poisoning, making malicious activities more scalable and less labor-intensive.

To counter these evolving threats, experts recommend proactive defensive strategies such as automated red teaming, zero-trust principles, strict identity controls, and human oversight.

Initial sessions have focused on understanding how to secure AI systems, with a final session scheduled for September 2 to explore how AI-driven attacks can bypass traditional defenses and how to build resilience.

A recent study highlights that AI-assisted code development can inadvertently introduce vulnerabilities, underscoring the need for human review, automated security tools, and developer training in AI security practices.

The OWASP warns about agentic AI executing sophisticated threats like memory poisoning and privilege escalation, emphasizing the importance of strong identity management and continuous monitoring.

NIST is hosting a series of virtual workshops to examine AI's dual role in cybersecurity—as both a defensive tool and a potential weapon for attackers.

The upcoming NIST workshop is vital for preparing defenses against current and future AI-enabled cyber threats, stressing the need for proactive and adaptive security measures.

Adversaries are experimenting with agentic AI capable of autonomously executing complex, multi-step attacks, with recent simulations showing breaches happening in roughly 25 minutes—up to 100 times faster than traditional methods.