HiddenLayer, an Austin-based AI security company, will contribute to the U.S. Department of Energy's $60 million Prometheus initiative, led by Idaho National Laboratory, to secure AI used in nuclear e…
A Nature study published in July 2026 found that four large reasoning models achieved a 97.14% overall jailbreak success rate against nine frontier AI models from OpenAI, Anthropic, and Google, using …
AEGIS, an AI security platform, now covers all seven major attack vectors—prompt injection, agent hijacking, MCP attacks, RAG poisoning, supply chain, model extraction, and vector database breaches—wi…
A maximum-severity vulnerability, tracked as CVE-2026-45829, has been discovered in the Python FastAPI version of the open-source AI database ChromaDB, allowing unauthenticated attackers to execute ar…