U-Rec Security InsightsInsights from U-Rec on AI agent and robotics safety, AI security research, development, and implementation.https://u-rec.jp/enOpenAI's AI Standards Proposal: Four Questions for Companieshttps://u-rec.jp/en/insights/openai-ai-standards/https://u-rec.jp/en/insights/openai-ai-standards/Four questions for AI deployment decisions, prompted by OpenAI's standards proposal. What evidence to keep for evaluations, approvals and incident reporting.Wed, 23 Sep 2026 16:33:00 GMTGuidanceAI governanceAI securityAI evaluationRecursive self-improvement (RSI)OpenAIWhen AI Development Tools Become the Entry Point: Settings and Permissions to Reviewhttps://u-rec.jp/en/insights/ai-coding-tools-supply-chain/https://u-rec.jp/en/insights/ai-coding-tools-supply-chain/Google's threat intelligence team reported attacks on AI coding assistants and MCP servers. What to check across external tools, credentials, and execution permissions in a development environment.Tue, 22 Sep 2026 11:54:36 GMTThreat AnalysisAI codingMCPSupply chainDevelopment environment securityThreat Analysis for AI Systems with Image Inputshttps://u-rec.jp/en/insights/image-injection/https://u-rec.jp/en/insights/image-injection/Adversarial inputs to image-recognition AI and image-borne prompt injection against multimodal AI: how they differ from ordinary variation, how to build a threat model, and what to measure.Tue, 22 Sep 2026 11:54:36 GMTThreat AnalysisImage recognitionThreat analysisMultimodal AIAdversarial inputsSecurity Design for Physical AIhttps://u-rec.jp/en/insights/physical-ai-security/https://u-rec.jp/en/insights/physical-ai-security/Safety and security for Physical AI: separating proposal, authorization, and execution, protecting ROS 2 communication, validating commands, and deciding fault responses at the design stage.Tue, 22 Sep 2026 11:54:36 GMTResearchPhysical AIRoboticsSecurity designROS 2Audit Log Design for Robotics Systemshttps://u-rec.jp/en/insights/ros2-audit-log/https://u-rec.jp/en/insights/ros2-audit-log/What a robotics audit log should record: how it differs from ROS 2 debug logs and rosbag, the fields linking decision, authorization, and execution, time and ordering, tamper detection, retention.Tue, 22 Sep 2026 11:54:36 GMTEngineeringRoboticsAudit logsROS 2ObservabilityAssessing Person-Detector Misses: An AI Security Perspectivehttps://u-rec.jp/en/insights/mask-extent-texture-study/https://u-rec.jp/en/insights/mask-extent-texture-study/For teams deploying or improving person-detection systems. Our study measures public detectors' detection rates on the same scenes under conditions that include textures built to suppress detection. We show how its method supports an AI security evaluation that grounds threat scoping, verification priorities, and countermeasure decisions in evidence.Sat, 19 Sep 2026 00:00:00 GMTResearchperson detectionadversarial textureAI securitydigital compositingevaluation methodologyrobustness evaluation