Sovereign AI Foundation | Eclipse Foundation

> An Eclipse Foundation Industry Collaboration
AI sovereigntystarts in the open.
AI increasingly shapes critical operations and decisions, yet many organisations rely on systems they cannot fully inspect, adapt, or govern. The Sovereign AI Foundation brings members and open source communities together, combining real-world experience with technical expertise to understand those dependencies, evaluate open source alternatives, and develop practical guidance for retaining choice and control.
// Open to all eligible Eclipse Foundation member organisations.
The strategic risk of AI dependency
The choices organisations make about models, data, infrastructure, and providers can shape their costs, resilience, compliance obligations, and freedom to change direction. Many of the models, platforms, and infrastructure underpinning AI are controlled by a relatively small number of providers, making the resulting dependencies harder to understand or change.These dependencies can narrow future options. They may limit where data is processed, which technologies can be used, whether workloads can be moved, and how organisations can respond when requirements, costs, or provider terms change.This is why AI sovereignty matters. In practical terms, it is about preserving four essential capabilities:
Maintain continuity, security, and performance when connectivity, providers, or operating conditions change.
Determine where sensitive data is processed and how AI systems are deployed at the edge, on-premises, in the cloud, or across multiple environments.
Evaluate, combine, replace, or adapt models, infrastructure, and providers without having to rebuild entire systems.
Understand, explain, and challenge AI-supported decisions, particularly when they affect people, operations, or public services.
AI sovereignty is an organisation’s ability to make informed choices and retain meaningful control over the AI systems on which it depends. Organisations can use technologies from many sources and collaborate across a global ecosystem while preserving control over their data, deployment choices, critical dependencies, and ability to change direction.
Open source provides the foundationOpen source makes critical technologies easier to inspect, evaluate, adapt, combine, and replace. This expands choice, supports viable alternatives, and reduces unnecessary dependency on any single provider.
Access to source code or model weights alone, however, does not guarantee sovereignty. Organisations also need portable architectures, clear licensing and provenance, operational expertise, sustainable communities, and trusted governance.
The Sovereign AI Foundation helps members understand these factors, compare available approaches, and apply them to real-world AI decisions.
The Sovereign AI Foundation gives members a structured way to learn from one another, assess technologies and dependencies, and engage with open source communities relevant to their needs.
Provides the governance, legal framework, and operational support that enable organisations and project communities to collaborate in a vendor-neutral environment.
Bring real-world use cases, technology dependencies, and operational requirements. Members learn from one another, identify shared needs, assess available options, and develop practical guidance.
Develop and maintain open source AI technologies under established project governance. Members can evaluate and adopt relevant projects, contribute expertise and requirements, and participate directly in their communities.
The Sovereign AI Foundation does not govern projects or determine their technical roadmaps. Projects remain independent, while members gain a clear path to explore their technologies, share feedback, and contribute where their interests align.
-> Explore AI at the Eclipse Foundation
Members focus on six areas that help them understand their options, learn from real-world experience, and make better-informed AI decisions:
Build a shared view of how organisations are evaluating, piloting, and deploying open source AI, including common challenges and priorities.
Evaluate developments across models, infrastructure, agentic systems, and evaluation methods, including the dependencies, limitations, and operational risks they may introduce.
Explore how projects align with member requirements and identify opportunities for adoption, contribution, and collaboration.
Compare lessons from real-world deployments and identify requirements that members and project communities can address together.
Create white papers, blueprints, landscape analyses, and reference materials that help members make informed decisions.
Turn findings from Eclipse Foundation research initiatives, including EU-funded projects, into insights members can apply.
Members can evaluate projects across the Eclipse Foundation portfolio against their technical, operational, and strategic requirements. They can also engage directly with project communities, contribute expertise and feedback, and identify gaps and opportunities for further collaboration.
Open technologies for building AI-enabled development tools and integrated development environments, with flexibility over models, integrations, and user experience.
Technology for running AI coding agents in isolated environments with defined security and operational policies.
Explore Eclipse Enclave –>
An open platform for evaluating AI model capabilities, safety, and cybersecurity.
Explore Eclipse PanEval –>
An open platform for developing and managing multi-agent systems.
Explore Eclipse LMOS –>
A vendor-neutral registry for extensions used by open source and AI-enabled development tools.
Explore Open VSX Registry –>
These projects address different layers and use cases and remain independently governed. Together with the wider Eclipse Foundation portfolio, they provide modular technologies that organisations can evaluate, combine, and adapt without committing to a single integrated platform.
Partic