https://threatnotes.dev Security Research · Detection Engineering · Threat Hunting · AI Security en-us https://threatnotes.dev/post/ai-gateway-telemetry-in-the-soc.html https://threatnotes.dev/post/ai-gateway-telemetry-in-the-soc.html Tue, 21 Jul 2026 00:00:00 +0000 Part 2 of the series. A working demonstration that turns AI Gateway (LLM Guard) telemetry into LogRhythm detections (prompt injection, credential exposure, PII/DLP, prohibited topics, malicious URLs, and token abuse), each blocked at the gateway and made investigable in the SIEM. https://threatnotes.dev/post/ai-gateways-ai-agents-layered-security.html https://threatnotes.dev/post/ai-gateways-ai-agents-layered-security.html Thu, 16 Jul 2026 00:00:00 +0000 AI Agent monitoring shows what an agent did, but an agent only executes a decision the AI Model already made. A complete AI security strategy secures every layer, starting with the decision itself at the AI Gateway. Here's how the layers fit together.