Analysis ·
AI agents: the platform becomes the key architecture question
Original title, in French: “Agents IA : la plateforme devient le vrai sujet d'architecture”
- Publication
- Journal du Net (France)
- Author
- Michel Hubert
- Published on
- 1 October 2026 · read in the page structured data
Summary
The article analyses recent changes by Microsoft and GitHub that enable the large-scale deployment of AI agents in enterprises. Since 24 September, Routines in Microsoft Foundry Agent Service make it possible to trigger agents on a schedule or in response to an event, with identity and permissions management. Microsoft also proposes separating agent governance (in Foundry) from execution (in isolated Azure sandboxes). GitHub adds local sandboxing for Copilot and OpenTelemetry trace export to monitor agent behaviour, as well as a breakdown of pull request review time. The arrival of new models (GPT-6 Sol, Luna, Astra) in Foundry widens the choice according to the task. The article concludes that the challenge for platform teams is now to define a framework for the governance, identity, execution and evaluation of AI agents.
Summary based on the full text of the article.
Read the article on Journal du Net →Classification
- Country
- Worldwide scope
- Company size
- All sizes
- Industries
- Software vendors, IT services companies, telecoms
- Companies
- GitHub (supplier), Microsoft (supplier)
- Keywords
- AI agents, governance, platform architecture, microsoft foundry, github copilot, observability
Entry written by Marsanne Studio from the cited article, with the help of an AI model (claude-sonnet-5). The rights to the article and its illustrations belong to Journal du Net and its author. Entry translated from French by Claude Sonnet 5.5.