Skip to content
Tagged articles

security evaluation

5 articles · Page 1 of 1
Machine Heart
Machine Heart
Nov 13, 2026 · Artificial Intelligence

Pandora's Toolbox: Mapping Security Risks Across the AI Agent Lifecycle

Peking University and WeChat Security team present a lifecycle-based survey of generalist AI agent security, analyzing how risks propagate across provenance, orchestration, execution, and persistent feedback stages, covering attack surfaces, evaluation benchmarks, and defense layers.

AI agent securityattack surfacesdefense mechanisms
0 likes · 19 min read
Pandora's Toolbox: Mapping Security Risks Across the AI Agent Lifecycle
Architects' Tech Alliance
Architects' Tech Alliance
Oct 22, 2026 · Information Security

China Security Evaluation 2026 No.3: CPUs Reach Level III for First Time, Printer Controller Chips Debut

China's Information Security Evaluation Center released its 2026 third batch of security and reliability results, with 7 CPUs (3 at Level III, 4 at Level II), 2 AI chips, 4 printer controller chips, 2 server OSes, and 3 centralized databases certified — marking the first appearance of CPUs at Level III and first publication of printer controller chip results.

AscendGalaxy KylinKunpeng
0 likes · 4 min read
China Security Evaluation 2026 No.3: CPUs Reach Level III for First Time, Printer Controller Chips Debut
Machine Heart
Machine Heart
Aug 8, 2026 · Artificial Intelligence

Why Anthropic Says Claude Code’s Auto Mode Is Safer After Testing 1,000 Users

Anthropic’s new default Auto Mode for Claude Code uses a dedicated classifier that caught 89% of dangerous commands versus 14% for manual approval, a study of 1,053 paid testers showed equal or better safety, fewer harmful actions, and zero successful attacks on Claude models compared with competing systems.

AI safetyAgent ToolsAnthropic
0 likes · 11 min read
Why Anthropic Says Claude Code’s Auto Mode Is Safer After Testing 1,000 Users
AntTech
AntTech
May 25, 2026 · Artificial Intelligence

Ant Group and Five Universities Unveil Agent3σ: A Multi‑Layer AI Agent Security Evaluation Platform

As AI agents move from simple Q&A to tool use and real‑world actions, Ant Group and five leading universities launch the open‑source Agent3σ platform, offering a three‑tier, 7‑dimension risk framework and concrete metrics to assess agents' safety across static, simulated, and live environments.

AI agentAgent3σMulti‑Layer Testing
0 likes · 9 min read
Ant Group and Five Universities Unveil Agent3σ: A Multi‑Layer AI Agent Security Evaluation Platform
Tencent Technical Engineering
Tencent Technical Engineering
Jul 16, 2025 · Artificial Intelligence

Introducing A.S.E: The First Project‑Level AI Code Generation Security Evaluation Framework

The A.S.E (AI Code Generation Security Evaluation) framework provides a comprehensive, project‑level benchmark for assessing the safety, quality, and stability of AI‑generated code across multiple languages and vulnerability types, helping developers and researchers evaluate and improve large language model coding assistants.

AI code generationopen-source benchmarksecurity evaluation
0 likes · 7 min read
Introducing A.S.E: The First Project‑Level AI Code Generation Security Evaluation Framework