- Google Cloud launches new AI Protection security suite
- Offering identifies, assesses, and protects AI assets for vulnerabilities
- Even more security features are coming soon
Google Cloud has launched AI Protection, a suite of security features designed to mitigate risks across AI workloads and data, regardless of the platform used.
The new offering will give businesses a centralized view of their AI standing, allowing them to manage the risks and spot threats before they become a concern.
“As AI use increases, security remains a top concern, and we often hear that organizations are worried about risks that can come with rapid adoption,” noted Archana Ramamoorthy, Senior Director, Product Management, Google Cloud Security. “Google Cloud is committed to helping our customers confidently build and deploy AI in a secure, compliant, and private manner.”
Boosted protection for AI workloads
AI Protection will be built into Security Command Center (SCC), providing a centralized AI security management system alongside other cloud risks.
Among the core capabilities of the new platform are AI Inventory Discovery (identifies and assesses AI assets for vulnerabilities), AI Asset Security (implements controls, policies, and guardrails to secure AI resources), and Threat Management (offers detection, investigation, and response mechanisms for AI-related threats).
Furthermore, Google Cloud explained that its Sensitive Data Protection (SDP) Enhancements now extend to Vertex AI datasets, enabling automatic discovery and classification of sensitive training and tuning data. After discovering sensitive data, AI Protection will use SCC’s virtual red teaming to identify potential attack paths on AI systems and suggest remediation steps.
Google Cloud also said Model Armor, a core capability of AI Protection, is now generally available. It is designed to protect against prompt injection and jailbreak attacks, data loss and malicious URLs, and offensive content. It can be integrated into applications via REST API, Apigee, and soon Vertex AI.
Finally, AI Protection will operationalize security intelligence and research from both Google and Mandiant to help defend AI systems.
Initial access attempts, privilege escalation, and persistence attempts for AI workloads can all be detected via SCC, while new detectors to AI Protection, based on the latest frontline intelligence, are “coming soon”. These will help identify and manage runtime threats such as foundational model hijacking.
Leave a comment