AI Security & Reliability
AI security and reliability protect AI systems and keep them operating as intended.
Follow vulnerabilities, access controls, supply-chain failures, and service outages. Coverage looks at what broke, how teams secured or restored affected systems, and which mitigations they can verify.
Latest coverage
Articles
NewsCheck Point says Daybreak cuts vulnerability research from hours to minutes
NewsTogether AI makes automatic GPU-cluster repair controls generally available
NewsSalesforce plans evaluation-driven shutdown controls for Agent Fabric
NewsMBZUAI study finds speech-model updates leak age and accent signals
NewsOpenAI resumes frontier-model training without live internet access
NewsCohere adds confidential computing to Model Vault
NewsOpenAI and Poland’s NASK sign AI safety and security agreement
NewsCrowdStrike documents four agentic-AI hacking campaigns
NewsPHP patches critical RCE flaws as Dutch government sites go offline
NewsOpenAI to require approval for model-access intermediaries
NewsOpenAI outlines Patch the Planet role in response to two critical PHP flaws
NewsMeta fixes bug that exposed WhatsApp chatbot instructions on help pages