Check Point Lakera Red is an AI-native red teaming solution that performs security assessments of generative AI applications. The product tests for vulnerabilities through three primary attack vectors: direct manipulation to extract sensitive data or force harmful content exposure, indirect manipulation through backdoor injection or persistent manipulation of data sources, and infrastructure attacks to identify unauthorized access or privilege escalation risks. The platform provides risk-based vulnerability management by prioritizing vulnerabilities based on potential impact and risk exposure. It delivers collaborative remediation guidance to Product, Security, and Engineering teams. The solution is supported by threat intelligence derived from Gandalf, a security game that has built a community of AI security researchers. Lakera Red conducts security and compliance risk assessments of AI systems used by organizations and provides expert recommendations for addressing identified vulnerabilities. The product aims to help organizations accelerate AI readiness by translating vulnerability discoveries into actionable security measures.