Unamentis

Our commitment to keeping UnaMentis free and open source isn't just a policy. It's something we intend to protect permanently. Our long-term goal is to establish a formal nonprofit structure that ensures this project can never be sold, closed, or made proprietary.

Until that structure is in place, we've partnered with Open Collective Europe, a Belgian nonprofit that provides fiscal hosting for open source projects. This allows us to receive donations transparently while maintaining our independence.

The initial motivation for this project was born out of my personal discovery of how incredibly valuable—and how uniquely powerful—personal learning with AI can be. That discovery happened about eight months ago, and it was genuinely game-changing for me.

I started with ChatGPT Advanced Voice Mode and began feeding it topics, asking for detailed lectures. I spent a lot of time barging in with questions, going on tangents, and pulling threads as far as I wanted. It quickly became one of the most complete, efficient, and enjoyable educational experiences I’ve ever had.

Over the next few months, though, I kept running into issues any time I tried to follow that same pattern consistently. Sometimes it worked reasonably well. Other times—regardless of which AI tool I used—trying to do this kind of learning in voice mode was basically useless.

This finally came to a head when I had an extremely important use case: a set of topics I urgently needed foundational knowledge in. I went to use this same approach to solve that problem and got nowhere. That’s when I decided it was time to build a solution specifically for myself to meet this need.

As I started developing and refining what I wanted this to be, the scope expanded quickly—within just a few days it became obvious that I needed to run with this and take it as far as I could. I do see real business potential here, but I have to emphasize something: the primary and uncompromising focus for me (and anyone joining me on this project) is that the foundation app and the service supporting it will be free and open source forever. I will go to great lengths to ensure that remains permanently true.

Our plan is to bring this app to other platforms, expand it into as many languages as we possibly can, and keep pushing what it can do and what curriculums it can offer. We’re already connected into a couple of content sources (including MIT and a specific K–12 source), but ultimately we want people to be able to learn not only from standardized and well-respected curriculum, but also to create their own—or upload resources they already have and have those materials digested into a real curriculum they can work through.

And while the experience is voice-first and foremost, we’ve already added support for images, so maps, diagrams, and pictures can be part of the curriculum and show up in the transcript. On devices like an iPad or other tablets, there’s even more room to explore those visuals while listening.

The key feature, though, is the full round-trip interaction with AI while learning: at any point you can stop, barge in, ask questions, go on tangents, explore ideas, ask for examples or illustrations—whatever you need—and then return right back to where you left off in the curriculum you’re working through.