Microsoft Details New Security Safeguards for Generative AI Models on Azure AI Foundry

Microsoft describes security safeguards for generative AI models on Azure AI Foundry, warning that threat actors could embed malicious code or backdoors in models and turn them into supply-chain infection vectors. The company details a multi-stage pre-release scanning process (malware analysis, vulnerability assessment, backdoor detection, integrity checks) and manual review/red-team testing for high-visibility models, and advises customers to verify model-scan indicators and apply zero-trust governance when deploying third-party models.