Open Source Initiative
In plain wordsA nonprofit group that decides what counts as 'open source' software and, in 2024, published a definition of what makes an AI system genuinely open source.
A US nonprofit that maintains the Open Source Definition — ten criteria including free redistribution, availability of source code, permission to make derived works, and no discrimination against persons, groups or fields of endeavor — and approves software licenses that meet it, such as MIT, Apache 2.0 and the GNU General Public License. In AI it is best known for the Open Source AI Definition (OSAID) 1.0, released on 28 October 2024 after a two-year public consultation. The OSAID asks that anyone be free to use, study, modify and share the system, and that they receive what is needed to modify it in practice: the model parametersweights, n.: The learned numbers inside a neural network that set how strongly each part influences the next. The finished set of weights is, in effect, the trained model itself., all the code for training and running it, and a description of the training data ('data information') full enough that an expert could recreate a comparable system — though not necessarily the data itself. The OSI has argued that LlamaLlama, prop. n.: A family of AI language models from Meta, first released in 2023, whose trained settings can be downloaded and run by anyone, making them the foundation for thousands of other AI projects. and other models released with use restrictions are not open source, and its definition is the main reference point in disputes over open-source AIopen-source AI, n.: AI released, ideally with its code and details of its training data and not just the finished model, so that anyone may use, inspect, alter, share and even rebuild it. How open is open enough is debated. and open weightsopen-weight model, n.: An AI model whose trained internal numbers are published for anyone to download and run on their own computers, even if the data and code used to make it stay private..
Because the license restricts certain users and uses, the OSI does not count the model as open source.
Origin
Founded in February 1998 by Bruce Perens and Eric S. Raymond, weeks after Netscape announced it would publish its browser's source code and after Christine Peterson proposed 'open source' as an alternative to 'free software'. The Open Source Definition was adapted from Perens's Debian Free Software Guidelines (1997). The OSI began work on an AI definition in 2022.
Usage note
The OSAID drew criticism from two sides: some open-source advocates said that not requiring training data lets opaque models qualify, while Meta, whose Llama models do not qualify, said no single definition could capture today's AI. Few prominent models meet it; those cited as conforming include EleutherAIEleutherAI, prop. n.: An open-science research group, begun by online volunteers in 2020, that released some of the first large language models anyone could download, along with the text collections used to train them.'s Pythia and Ai2Allen Institute for AI, prop. n.: Paul Allen, a co-founder of Microsoft, created this nonprofit Seattle AI lab in 2014. It is known for releasing models together with everything used to build them, and for the Semantic Scholar paper search engine.'s OLMo.
Related
- open-source AIopen-source AI, n.: AI released, ideally with its code and details of its training data and not just the finished model, so that anyone may use, inspect, alter, share and even rebuild it. How open is open enough is debated.
- open-weight modelopen-weight model, n.: An AI model whose trained internal numbers are published for anyone to download and run on their own computers, even if the data and code used to make it stay private.
- model licensemodel license, n.: The legal terms under which an AI model may be downloaded and used — for example, whether businesses may use it, whether changes may be shared, and which uses are banned.
- LlamaLlama, prop. n.: A family of AI language models from Meta, first released in 2023, whose trained settings can be downloaded and run by anyone, making them the foundation for thousands of other AI projects.
- data licensingdata licensing, n.: The legal permissions that say who may use a dataset or a body of content, for what purposes and on what terms — including whether it may be used to train AI.
- Hugging FaceHugging Face, prop. n.: A company whose website is the main public library for sharing AI models and datasets — often called the GitHub of machine learning — and which makes widely used free AI software.
Sources
- Open Source Initiative (2024). The Open Source AI Definition 1.0. opensource.org, 28 October 2024.
- Open Source Initiative. The Open Source Definition (version 1.9). opensource.org.
Mentioned in
gpt-ossgpt-oss, prop. n.: Two free-to-download reasoning models released by OpenAI in August 2025 — its first downloadable language models since 2019 — that anyone may run, modify and use commercially on their own hardware.