AgentCore Gateway combines Cedar policies for static access control with Lambda interceptors for dynamic validation, enabling secure governance of LLM-based agents at scale.
EU high-risk guidelines are now operationally deployable, all 27 member states have AI sandboxes, Anthropic documents its sandbox architecture, and 500-million-token contracts reveal the new scaling reality. What IT leadership must address this week.
Three threads shaped May: the AI Omnibus and first high-risk guidelines from Brussels, Claude 4.8 with KPMG scaling as commercial proof, and a wave of supply-chain incidents from Nx-Console to axios — what began in May becomes operational in June.
The EU is collecting feedback until 23 June 2026 on the clarity and practical applicability of its guidelines for classifying high-risk AI systems under the AI Act.
Claude Opus 4.8 reduces hallucinations by approximately 75 percent by abstaining more frequently on uncertain questions instead of providing unfounded answers.
Anthropic isolates Claude agents through multi-layered sandboxes (gVisor, Seatbelt, Bubblewrap, VMs) with explicit boundaries for data access, filesystem, and egress control.
Open models are closing the gap to the frontier, but different benchmarking methods and evaluation frameworks make reliable performance comparisons between open and closed systems difficult.
