Welcome to the December 2025 edition of the Linux Foundation Newsletter.
Winter is nearly here, and the Linux Foundation open source ecosystem continues to break new ground. This month, we announced the formation of the Agentic AI Foundation (AAIF), bringing together critical open standards and frameworks - including Model Context Protocol (MCP), AGENTS.md and goose - for next‑gen AI agents under a neutral, community‑driven umbrella. We also saw continued growth in global collaboration, advances in infrastructure and AI tooling, and strategic developments across projects that are shaping the future of open technology. Thank you to all the contributors, maintainers, members, and staff driving this impact forward and have a wonderful holiday season.
Here are more of this month’s highlights:
Agentic AI Foundation (AAIF) Launches to Advance Open Standards for AI Agents Last week we announced the launch of the AAIF with contributions from Anthropic’s Model Context Protocol (MCP), Block’s goose agent framework, and OpenAI’s AGENTS.md, with membership support from AWS, Google, Microsoft, Bloomberg, Cloudflare, Cisco, and other leading organizations. The move sets the stage for shared standards and tools as agentic AI systems scale across industries.
New Linux Foundation Research Report: The State of Open Source Japan 2025 At this year’s Open Source Summit Japan, LF Research released The State of Open Source Japan 2025, with new data highlighting how strategic open source engagement accelerates business value and innovation in Japanese enterprises. The report sheds light on adoption trends, challenges with governance and skills, and opportunities for open collaboration in cloud, AI, and digital transformation initiatives.
Mitsubishi Electric Joins as Linux Foundation Gold Member at Open Source Summit Japan Another big announcement out of the Open Source Summit Japan - Mitsubishi Electric Corporation has become a Gold Member of the LF, expanding industry participation in open source development across embedded systems, industrial automation, and next‑gen connectivity.
Read more about this announcement in our press release here.
The Xen Project announced the release of Xen 4.21, delivering virtualization enhancements for maintainability, performance, and security to cloud and server workloads
The AgStack Foundation announced a strategic collaboration to embed OpenAgri’s portfolio of software into AgStack’s digital infrastructure ecosystem
What’s Next?
Explore the Agentic AI Foundation projects on GitHub and get involved with MCP, goose, and AGENTS.md as they define the future of interoperable AI agents.
Over the past several years, LF Research has had the privilege of studying open source adoption across regions and industries worldwide. What consistently stands out about Japan is not hesitation, but intentionality. Japanese organizations are thoughtful, exacting, and deeply pragmatic in how they adopt technology, and our latest report, The State of Open Source Japan 2025, shows that this approach is paying off in measurable business value, even as important gaps remain. Last week in Tokyo I had the opportunity to share these findings with the attendees of Open Source Summit Japan, AI_Dev, and Automotive Linux Summit. Here are a few of the highlights for those who couldn’t join us in person.
Open source and open collaboration communities face ongoing threats from “non-practicing entities” (NPEs, also sometimes called “patent trolls”). The United States Patent and Trademark Office (USPTO) has proposed new procedural rule changes that would benefit NPEs, by making it harder to defend against NPEs who assert weak patents.
Artificial intelligence is reshaping economic systems at a pace we have rarely seen in modern technological history. Every sector—from finance to healthcare to manufacturing—is scrambling to understand how to harness AI safely, efficiently, and competitively. Yet amid the excitement, a crucial part of the story has been missing. Specifically, understanding the role that open models play in the AI economy, and how much value is being left on the table when organizations overlook open alternatives, are two topics requiring a closer look.
Welcome to the November 2025 edition of the Linux Foundation Newsletter.
As we move toward year‑end, open source activity at the Linux Foundation (LF) remains at full throttle. In the past month, we welcomed major new projects, strengthened our AI‑and‑infrastructure portfolio, and reinforced our global collaboration model across security, research, and innovation. A huge thank you to all contributors, maintainers, members and staff who keep this momentum going!
Here are more of this month’s highlights:
Valkey 9.0 Delivers Next‑Gen Performance at Scale The open‑source key‑value database project announced version 9.0 this month. This release introduces atomic slot migration, multiple databases in cluster mode, hash‑field expiration, and benchmarks showing support for over 1 billion requests per second across 2,000 nodes.
Read more about the latest version of Valkey in Diginomica
Fluxnova Launches Under FINOS to Orchestrate Financial Workflows The Fintech Open Source Foundation (FINOS) announced Fluxnova in partnership with Fidelity Investments, NatWest Group, Bank of Montreal, Deutsche Bank and Capital One. Fluxnova, a fork of Camunda 7, is an open orchestration platform enabling audit‑ready workflows, visual process models and process traceability in heavily regulated financial services environments.
Read more about what makes this platform so critical to the ecosystem in the SD Times
Overture Maps Foundation Names New Executive Director and Lands on Fast Company’s 2025 Next Big Things in Tech List William Mortenson has joined the Overture Maps Foundation as its new Executive Director. Mortenson brings more than 25 years of geospatial leadership and begins guiding Overture’s next phase of open map‑data growth, interoperability and adoption. The project also saw major industry recognition this month, landing a spot on the coveted ‘2025 Next Big Things in Tech’ list by Fast Company.
PyTorch Foundation Welcomes “Ray” to Deliver a Unified Open Source AI Compute Stack The PyTorch Foundation announced its latest hosted project: Ray, a widely adopted distributed computing framework that enables scaling AI workloads from a single machine to thousands of nodes. Ray now joins PyTorch and vLLM under the PyTorch Foundation umbrella, reinforcing the open‑source AI stack.
Major Infrastructure & Edge Release: StarlingX 11.0 The open‑source cloud infrastructure project hosted by the Open Infrastructure Foundation released version 11.0, bringing enhanced edge‑security, IPv4‑exhaustion mitigation, IPsec pod‑to‑pod encryption and stronger rollback support for complex multi‑cluster deployments.
Read more about the new feature and optimization updates in Network World
What’s Next?
Visit the Ray, Valkey, and Fluxnova GitHub repositories to explore updates and join the community.
The recent GOSIM AI Vision Forum in Hangzhou crystallized the central paradox of artificial intelligence: how to harness its immense potential while mitigating its considerable risks. AI is already augmenting our capacity for knowledge work by simplifying discovery, synthesis, and translation, as well as automating routine tasks. This automation frees individuals to engage in more meaningful and creative endeavors. Yet, these advancements are shadowed by urgent challenges, including equitable access, value-aligned governance, and protecting our social fabric from harm. As Dr. Michael Yuan highlights in the foreword of the recent Linux Foundation report Global Cooperation for Human-Centered AI, aligning AI with human values is crucial, but equally important is preparing humans to collaborate effectively with AI—a shift that requires us to evolve our fundamental understanding of work, education, and creativity.
In August, I had the honour of attending the APEC 2025 Global Digital and AI Forum’s AI & Digital Ministerial Meeting in Incheon, South Korea. I was thrilled to have the opportunity to present findings from the first phase of our research in collaboration with Meta, and to join a group of esteemed panelists discussing the numerous pathways for open source AI to transform the region’s 21 member states in different capacities.
This week in New York City at the Open Source in Finance Forum we marked an exciting milestone. This is the fifth consecutive year that Linux Foundation Research and FINOS have collaborated on the State of Open Source in Financial Services Report, and the insights from this year's study are not only impactful for the financial services sector at large, they extend far beyond it.
Welcome to the October 2025 edition of the Linux Foundation Newsletter.
Autumn is upon us and open source innovation shows no signs of slowing. Over the past month, the Linux Foundation welcomed new projects, celebrated major project milestones, and advanced our mission of enabling open collaboration across industries. Here are more of this month’s highlights:
React Foundation Launches Under the Linux Foundation The Linux Foundation announced the formation of the React Foundation, a new home for React, React Native, and supporting projects to thrive under neutral, open governance. Contributed by Meta and backed by industry leaders, the foundation aims to support the long-term sustainability of one of the world’s most popular front-end frameworks.
Disney Research, NVIDIA, and Google DeepMind Contribute Newton Physics Engine A new project, Newton, has been contributed to the Linux Foundation by Disney Research, DeepMind, and NVIDIA. Newton is a GPU-accelerated physics engine designed for robotic simulation and reinforcement learning—bridging the gap between virtual training and real world deployment.
LF Decentralized Trust (LFDT) Celebrates One Year LF Decentralized Trust marked its one-year anniversary with key milestones, including new members, community growth, and the move of Hiero, the distributed ledger technology powering the Hedera network, to graduated project status. The initiative continues to drive forward open source solutions for tokenized assets, verifiable credentials, decentralized identity, and public trust.
In this second report in our series on the economic value of open source AI, we reviewed the technology’s impact in Africa, the Middle East, and Türkiye (AMET). Drawing on evidence from industry and academia, the study reveals strong adoption and investment trends, enormous economic potential, and transformational workforce and sector impacts. Many of the themes from our global study ring true in AMET as well, alongside some findings that are unique to this region.
