Safety & security

Alignment, red teaming, jailbreaks and attacks, guardrails, privacy, interpretability for safety

300 articles · page 4 of 4

All Agents Language models Reinforcement learning Vision Generative media Speech & audio Robotics & embodied AI Safety & security Evaluation & benchmarks Efficiency & systems Code & software AI for science & health Theory

Viewgrid list compact

Dedicated intent tools expose shifts toward harmful agent behavior

Reinforcement learning strengthens policy invocation for agent safety judgments

Semantic scores overstate language-model safety in air traffic control

Medical AI models often give right answers with disconnected reasoning

Physiological grounding makes clinical language-model recommendations safer in controlled tests

Step-level checks curb unsafe agent actions with little utility loss

Masking untrusted web content to secure agents against prompt injection

Persistent memory in AI agents can be stealthily poisoned via a single email

Persistent-codebase AI agents vulnerable to distributed, gradual attacks

Automated safety testing reveals 93.9% attack success rate across four agent frameworks

AI progress may continue past AGI toward superintelligence, not stop

First measurement study reveals widespread authentication flaws in remote MCP servers