Introduction:

Recently, Forrester, a leading global research and advisory firm, released its report titled "Navigate The AI-Native Cloud Security Ecosystem In China." The report systematically outlines the trends and vendor landscape of China’s AI-native cloud security ecosystem, including Alibaba Cloud among leading Public Cloud Platform (PCP) vendors embedding AI-enabled security into integrated cloud platforms.

With its comprehensive AI Agent security product portfolio and deep scenario coverage, Alibaba Cloud was listed in seven example customer cases defined in the Forrester report, which Alibaba Cloud believes fully demonstrates its leading position in China’s AI-native cloud security domain.

From Cloud-Native to AI-Native: A Fundamental Paradigm Shift in Security

Forrester notes in the report that China is rapidly shifting from cloud-native to AI-native architectures.

Traditional static infrastructure hardening and perimeter controls are no longer sufficient to address the unpredictability, expanding attack surfaces, and stringent data compliance requirements brought by generative AI and intelligent agents. Organizations urgently need to establish continuous, AI-driven intelligent security governance systems.

The report proposes that AI-native cloud security is built upon two pillars:

1. Cloud Security For AI: Building the cloud platform into a trusted AI foundation to defend against emerging threats such as model poisoning, prompt injection, and AI compute hijacking across four layers: data, model, application, and infrastructure.

2. AI For Cloud Security: Leveraging Large Language Models (LLMs) and AI Agents to enhance the intelligence of cloud security, enabling automated threat prediction and response from the infrastructure to the application layer.

Alibaba Cloud Security: Listed in Forrester's Seven Example Customer Cases

Alibaba Cloud's security product portfolio and real-world capabilities comprehensively cover all seven example customer cases of AI-native cloud security. Below are the seven example customer cases from Forrester’s report, as well as Alibaba Cloud’s input on how their platform helps with them.

Cloud Security For AI

1. Data Security: Through its Data Security Center (DSC) and Key Management Service (KMS), Alibaba Cloud accurately identifies and automatically de-sensitizes Personally Identifiable Information (PII) during complex business interactions involving massive volumes of policy images, text, and other data, significantly reducing the risk of privacy leaks during AI utilization. To prevent sensitive data leakage while using AI, Manulife-Sinochem deployed Alibaba Cloud's DSC to ensure strict access governance and strengthen compliance, reducing deployment costs and simplifying global compliance processes.

2. Model Security: Leveraging AI Guardrails and Web Application Firewall (WAF), Alibaba Cloud provides low-latency, streaming protection for LLMs. Through multimodal understanding and contextual analysis, it significantly improves the recall and accuracy of high-risk content detection, effectively blocking prompt injections and malicious exploits. SiliconFlow utilizes Alibaba Cloud's AI Guardrails for real-time model protection, reducing latency through streaming moderation and algorithm orchestration, which increased high-risk content detection recall by 8% and accuracy by 11%.

3. Application Security: Alibaba Cloud provides model-level security protection through LLM-oriented WAF. Combined with centralized AI asset management and automated vulnerability remediation capabilities, it safeguards the entire lifecycle of AI applications from development to runtime. A leading new energy vehicle (NEV) manufacturer embedded Alibaba Cloud's application security checks into its AI Agent and cloud-native CI/CD pipelines for autonomous driving R&D, automatically detecting and remediating over a thousand vulnerabilities.

4. Infrastructure Security: Alibaba Cloud provides AI/Agent-SPM (Security Posture Management) and AI/Agent-BOM (Bill of Materials) capabilities to achieve end-to-end AI Agent asset inventory. It also offers rigorous protection for core AI computing power, such as GPU computing clusters, to ensure infrastructure safety. Turbo AI utilizes Alibaba Cloud Security Center to build a real-time detection and automated response defense system, using CSPM (Cloud Security Posture Management) to ensure baseline security configuration compliance, while one-click vulnerability patching significantly reduces the risk of GPU instances being exploited.

AI For Cloud Security

1. Infrastructure Threat Prediction: Based on all-domain log ingestion and driven by intelligent models, Alibaba Cloud's Agentic SOC platform enables autonomous judgment and intelligent closed-loop response orchestration. It generates intuitive attack paths, automates root-cause analysis, and provides intelligent hardening recommendations, reducing tens of thousands of alerts to just over a dozen high-risk events. A large national joint-stock commercial bank deployed Alibaba Cloud’s AI-driven security posture management capabilities, shortening the average remediation time for critical vulnerabilities to just one day and completing high-risk attack path assessments within three days, substantially reducing configuration-related security incidents.

2. Workload Behavior Analysis: By deploying security agents in containers and GPU clusters combined with cloud-based machine learning engines, Alibaba Cloud performs real-time anomaly baseline comparison on files, processes, and network activities to accurately capture hidden trojans and backdoors. To ensure the safe and stable operation of its GPU clusters used for model training and inference, Moonshot AI leverages AI-enhanced Alibaba Cloud security to detect cryptomining, backdoors, and trojans, achieving unified security management across multi-cloud and on-premises GPU clusters.

3. Application Security Automation: Alibaba Cloud provides AI-enhanced Bot Management (Anti-Bot) and API security solutions. During high-traffic enterprise marketing campaigns, it dynamically identifies and blocks malicious crawlers, vastly improving the stability of core business systems and the end-user experience. During a high-traffic marketing campaign, a software development enterprise utilized Alibaba Cloud's LLM-driven Bot Management capabilities to dynamically identify and block illegal activities like malicious crawlers, reducing disruptions to normal business operations and significantly improving user experience.

Just as the Forrester report reveals, cloud security in the AI era is undergoing a profound genetic restructuring.

Alibaba Cloud is consistently committed to its dual-engine strategy of "Security for AI" and "AI for Security." From infrastructure to model applications, and from data protection to autonomous agent operations, Alibaba Cloud is delivering an intelligent security foundation for global enterprises—one that can be seamlessly embedded into business workflows and evolves continuously alongside AI.

Information in Forrester publications is based on Forrester’s efforts to compile and analyze the best resources reasonably available to Forrester at any given time. Opinions reflect judgment at the time and are subject to change. This report is part of a broader collection of Forrester resources, including interactive models, frameworks, tools, data, and access to analyst guidance.