Artificial intelligence (AI) is increasing security risks in cloud systems as more companies deploy AI tools and workloads, according to the 2025 State of Cloud Security Report from Palo Alto Networks.
The report from the cybersecurity solutions provider said the rapid use of enterprise AI is expanding what security teams need to protect, often faster than existing controls can keep up. Based on a survey of more than 2,800 security executives and practitioners across 10 countries, the study found that 99% of respondents experienced at least one attack targeting their AI systems in the past year.
As organizations move AI workloads to the cloud, attackers are increasingly focusing on cloud infrastructure itself. The report noted that AI-assisted software development, sometimes called GenAI-assisted or “vibe” coding, is now used by 99% of respondents. While this speeds up development, it also creates more insecure code. Among teams that release code every week, only 18% said they can fix security flaws at the same pace, allowing risks to accumulate.
“Our research confirms that traditional approaches to cloud security are inadequate, leaving security teams to fight machine-speed threats with fragmented tools and slow, manual fix cycles,” said Elad Koren, VP of Product Management, Cortex. “Teams need more than just dashboards highlighting risks they can never burn down; they must transform with an agentic-first platform that spans code to cloud to SOC to finally operate faster than the adversary.”
The report also found about the growing role of application programming interfaces, or APIs. API attacks grew by 41% as AI systems rely heavily on APIs to function. This has turned APIs into a common entry point for attackers, according to the report.
Identity and access management also remains a major weakness. About 53% of respondents cited loose access controls as a major concern, making it easier for attackers to steal credentials and move data out of systems. Also, 28% pointed to unrestricted network access between cloud workloads, which can allow attackers to move across systems once they gain an initial foothold.
Palo Alto Networks also highlighted operational challenges. Organizations now manage an average of 17 cloud security tools from five vendors. This tool sprawl creates blind spots and slows response times. As a result, 97% of respondents said consolidating cloud security tools is a priority.
Gaps between cloud security teams and security operations centers, or SOCs, further delay response. About 30% of teams said it takes more than a full day to resolve an incident. Almost 9 in 10 respondents said cloud, application, and SOC security need to be fully integrated to be effective.
The report concluded that as attackers use AI to speed up attacks, organizations need security approaches that combine risk reduction and incident response across cloud environments.