The Agentic AI Foundation, hosted by The Linux Foundation, has published the official conference schedule for AGNTCon + MCPCon Japan 2026, taking place September 10-11, 2026, at Belle Salle Shibuya Garden in Tokyo. Industry leaders and open source innovators from SoftBank, Hitachi, NTT, Google, Microsoft, Anthropic, and more will gather to explore the standards, infrastructure, security and governance needed to operate AI agents in production.
The two-day technical program focuses on practical realities of agentic AI adoption: how teams design systems that can act reliably, connect safely to tools and enterprise data, coordinate across agents, and operate under real world constraints.
As enterprises across Japan move AI agents into systems their businesses depend on, the questions get harder including where to draw trust boundaries when multiple agents interact, how to enforce approval gates and audit trails, and how to verify an agent's output. The program was built around these questions, with more than 40 sessions providing real world evidence, honest trade offs, and lessons practitioners can put to work immediately.
"The rapid adoption of open agent frameworks and protocols across Japan underscores a clear demand for reliable, secure, and interoperable AI systems," said Mazin Gilbert, Executive Director of the Agentic AI Foundation. "As developers move from experimental prototypes to mission-critical systems, this conference showcases the hard technical work happening across Asia-Pacific, from tool integration and sandboxing to multi-agent coordination, evaluation and enterprise governance."
Highlights at AGNTCon + MCPCon Japan 2026 include Designing Trustworthy Agent Systems, Building the Open Agentic Stack, and Evaluating Agents Before Production.
Designing Trustworthy Agent Systems explores practical approaches to trust, oversight, and control in multi-agent environments as agents move closer to enterprise data and operational workflows. Session highlight: Designing Trust Boundaries for Agent-to-Agent Systems: Lessons Learned from a Technical PoC – Ryuji Iijima, SoftBank Corp.
Building the Open Agentic Stack covers how teams combine open source components, orchestration layers, authentication systems, observability tools, and emerging multi-agent patterns into complete architectures. Session highlight: What's Missing in the Open Agentic Stack: Lessons From OSS Integration – Tatsuya Sato and Satoshi Ito, Hitachi, Ltd.
Evaluating Agents Before Production addresses testing routing behavior, tool use, prompt changes, regressions, and edge cases before systems reach production. Session highlight: From Clicks To Context: Building an Open-Source Evaluation Pipeline for AI Agents - Inês Bolaños, PagerDuty.
Sessions span nine tracks including Building Reliable Agent Systems, Multi-Agent and Distributed Systems, Agentic Engineering, and Interoperability, Protocols and Standards. MCPCon, the dedicated MCP track, offers deep coverage of the fast-moving MCP ecosystem, including authorization and identity, transport and task lifecycle specifications, running MCP infrastructure at scale, and a keynote by co-creator David Soria Parra, Member of Technical Staff at Anthropic.
Attendees will also have discounted access to dedicated pathways for the Model Context Protocol Associate Certification.
About the Agentic AI Foundation
The Agentic AI Foundation is the neutral home where the agentic stack is being built. With founding projects including MCP, goose, AGENTS.md, and agentgateway, AAIF governs the core standards and protocols that enable agents to operate interoperably across platforms. Through transparent governance and broad industry participation, AAIF is driving adoption and ensuring agentic AI infrastructure evolves openly, predictably, and at production scale.
TechIntelPro empowers B2B technology decision-makers with the latest industry trends, deep insights, leadership perspectives and actionable intelligence to drive both organizational and professional milestones, and thrive on the tech pulse.
