Mission
Education that works for every learner — without compromising their privacy.
Around 16% of European students identify as neurodivergent. Local-first AI is how we close the gap, without sending a single second of recorded audio to a server we don't control.
Why we exist
The lecture is not the bottleneck. The note-taking after it is.
Most students who fall behind in class do so not because the material was too hard, but because the surrounding work — taking legible notes, re-reading transcripts, structuring review materials — happens at a pace and in a format that doesn't suit how their brain processes information.
For neurodivergent students that gap is wider. For students whose first language isn't the language of instruction it widens further. For students using a poorly-supported accessibility tool tied to a cloud subscription they can't always afford, it widens again.
NeuroBridgeEDU compresses that gap. The lecture is captured, transcribed, and summarised on the student's own device, in a format they choose, in any of 35 European languages — so they spend their effort understanding the content, not fighting the workflow around it.
UN SDG alignment
Mapped to three Sustainable Development Goals.
We don't claim to solve education on our own. We do claim to move three specific UN goals measurably forward in the institutions that adopt us.
UN Sustainable Development Goal
Quality education
Inclusive, equitable, lifelong-learning opportunities for every student. We make the lecture itself adaptable so the learner doesn't have to do the adaptation work alone.
UN Sustainable Development Goal
Reduced inequalities
Local-first AI removes the historic prerequisites for accessibility tooling — institutional procurement, expensive subscriptions, internet bandwidth. The same software runs in a Galway lecture hall and a rural Romanian classroom.
UN Sustainable Development Goal
Industry, innovation and infrastructure
Built on European foundation models (Mistral, EuroLLM, Whisper) so European education isn't dependent on US hyperscaler infrastructure for its core literacy tooling.
Commitments
What we will and won't do, regardless of business pressure.
Mission statements drift unless they're enforceable. These four commitments are the ones we hold as load-bearing — the brief we'd refuse to take a deal that violated.
Commitment
On-device by default
Recordings, transcripts, and summaries stay on the user's machine. Cloud features, where they exist at all, are explicit opt-in with a clear data-flow disclosure.
Commitment
Open source as the floor
The core product is free and open source. Code is auditable. Institutional features sit on top of an inspectable foundation, not behind it.
Commitment
European AI sovereignty
Mistral AI, EuroLLM, and Whisper power the product. The data, the models, and the company are all European — and we intend to keep it that way.
Commitment
Accessibility is the brief
WCAG 2.2 AA is the floor we build to, not the ceiling we measure against. Reduced motion, high contrast, low-stim mode, and screen-reader semantics are first-class concerns, not bolt-ons.
Join the mission
Use it. Test it. Tell your institution about it.
The fastest way the mission scales is one classroom adopting it at a time.