This week on The Business of Open Source, I spoke with Stefano Maffulli, Executive Director of the Open Source Initiative, about the definition of open source and… the definition of open source AI. We recorded this episode on-site at All Things Open, so there’s a little bit of background noise. 

We talked about why OSI felt like it needed to develop a definition of open source AI, how “open source” is enforced, and the thought process behind the definition that the OSI ultimately published. 

We talked about open data quite a bit — different kinds of data, what kind of information and data is important to researchers and professionals in the AI space, and if there’s a way to include AI models that are trained on proprietary data in the definition of open source AI. 

If you are interested in open source AI, definitely check out this behind-the-scenes discussion of how, and why, this definition was published — and what the future likely holds for defining open source AI.