Key developments this week show multi‑agent systems moving from research labs into production‑grade platforms. OpenAI announced that its GPT‑6.1 model now supports multi‑agent orchestration in beta, adding long‑running work and tool‑driven orchestration to the Responses API. Anthropic followed…
Multi‑Agent Systems Accelerate with New Vendor Orchestration, Provenance Tools, and Safety Scrutiny
The latest week has seen a confluence of product rollouts, security incidents, and research breakthroughs that together mark a turning point for multi‑agent AI. OpenAI introduced a beta‑level multi‑agent capability in its GPT‑6.1 Responses API, enabling developers to…
Security and Trust Advances Shake Up Multi‑Agent AI Landscape
Recent developments highlight both the growing power and the emerging risks of multi‑agent AI systems. Researchers at Zenity Labs demonstrated that a single publicly exposed AI agent on Amazon Bedrock’s AgentCore could seize control of every other agent…