Google Cloud fully supports enterprises' security capabilities through an integrated security strategy for the AI era, from real-time threat response to an autonomous security control center.

At the 'Google Cloud Security Summit 2025' held online on the 20th, Google Cloud announced a number of new security features for safe innovation in the AI era.

This presentation focuses on an integrated security strategy encompassing the entire AI ecosystem, including protecting AI agents, establishing an autonomous security control center, and expanding expert consulting.

Google Cloud has enhanced AI protection capabilities in its Security Command Center, enhancing its ability to protect sensitive data and respond to real-time threats.

In particular, the security of AI interactions has been dramatically improved by providing a preview version of the following features: △Automatic detection of AI agent inventory, △Prompt injection and jailbreak prevention, and △Position control functions through Google Agentspace and Google Agent Builder.

Additionally, proactive detection capabilities based on Mandiant's threat intelligence are applied to the Security Command Center, enabling real-time identification and response to abnormal behavior of AI agents.It became a crab.

Google Cloud also presented its vision for an autonomous security operation center (Agentic SOC).

Google SecOps builds a system where AI agents collaborate to analyze and respond to security alerts. Alert Investigation Agents autonomously perform event enrichment, CLI analysis, and process tree construction. This reduces manual work for security professionals and accelerates response times.

Mandiant's consulting services are also expanding to support corporate AI adoption. Through specialized consulting services such as risk-based AI governance, pre-deployment security guidelines, and AI threat modeling, the company will strengthen its security capabilities and help build a secure AI environment.

Google Cloud has also added various features to its unified security platform, Google Unified Security. These include SecOps Labs powered by Gemini AI, a unified SOAR data dashboard, and enhanced mobile security features for Chrome Enterprise.

Features for a trusted cloud environment have also been enhanced. Compliance Manager, data security posture management, autonomous IAM, and sensitive data protection have been expanded to include Vertex AI, BigQuery, and CloudSQL. Network security has also been further enhanced through Cloud NGFW and Cloud Armor Enterprise.

With this announcement, Google Cloud is redefining security standards for the AI era and empowering businesses to safely pursue innovation. The "Google Cloud Security Summit," targeting the Asia-Pacific region, will be held online on September 4th, with Korean subtitles available.