Uncategorised

Metagaming matters for training, evaluation, and oversight

Following up on our previous work on verbalized eval awareness:we are sharing a post investigating the emergence of metagaming reasoning in a frontier training run.Metagaming is a more general, and in our experience a more useful concept, than evaluati…

Agentic AI, AI Agents, AI Paper Summary, AI Shorts, Applications, Artificial Intelligence, Editors Pick, Language Model, Machine Learning, Promote, Sponsored, Staff, Tech News, Technology

Tsinghua and Ant Group Researchers Unveil a Five-Layer Lifecycle-Oriented Security Framework to Mitigate Autonomous LLM Agent Vulnerabilities in OpenClaw

Autonomous LLM agents like OpenClaw are shifting the paradigm from passive assistants to proactive entities capable of executing complex, long-horizon tasks through high-privilege system access. However, a security analysis research report from Tsinghua University and Ant Group reveals that OpenClaw’s ‘kernel-plugin’ architecture—anchored by a pi-coding-agent serving as the Minimal Trusted Computing Base (TCB)—is vulnerable to […]

The post Tsinghua and Ant Group Researchers Unveil a Five-Layer Lifecycle-Oriented Security Framework to Mitigate Autonomous LLM Agent Vulnerabilities in OpenClaw appeared first on MarkTechPost.

Scroll to Top