AI fuels escalating cloud security risks, Palo Alto Networks report reveals

Artificial intelligence has significantly amplified security risks in cloud environments, according to a new report from Palo Alto Networks. The findings detail a sharp escalation in threats as defenders struggle to keep pace with the speed and complexity of AI-driven attacks, according to Silicon Angle.

The Palo Alto Networks State of Cloud Security Report 2025 highlights that 99% of organizations utilize generative AI for software development, increasing code volume and introducing new risks like insecure configurations, vulnerable APIs, and prompt-based attacks. Nearly all surveyed organizations experienced attacks on AI systems, including data exfiltration via AI assistants and compromised credentials. API attacks saw a 41% year-over-year increase, becoming a critical pressure point as AI agents increasingly rely on them. Identity and access management also remains a weak link, with overly permissive practices cited as a major challenge. Data exfiltration is becoming more sophisticated, often disguised as legitimate activity, making detection difficult.

The report underscores the need to strengthen security earlier in the development lifecycle by optimizing pre-deployment controls and reducing incident response fragmentation. Implementing context-aware security gates within CI/CD pipelines is crucial. Elevating identity, permissions, and automation to top priorities, enforcing least-privilege access, and improving secret management are recommended to address the evolving threat landscape driven by accelerating cloud and AI adoption.

Source: Silicon Angle