AI’s next frontier: a kid-built robot that speaks a language only 10,000 people know

While Silicon Valley races toward artificial general intelligence, a more urgent AI challenge is unfolding in Michigan: giving voice to languages on the brink of extinction. SkoBot is a DIY robot that speaks Anishinaabemowin, trained on a corpus so small it would be dismissed by most commercial NLP pipelines—yet it's being built by the very kids who stand to lose their heritage tongue.

In Sault Ste. Marie, Michigan, a quiet educational revolution is taking shape on workbenches and classroom tables. Danielle Boyer, a member of the Sault Ste. Marie Tribe of Chippewa Indians, has invented SkoBot—a palm-sized AI-powered robot that speaks Anishinaabemowin, an Indigenous language spoken by fewer than 10,000 people in the United States. But the true innovation is not just the robot itself; it’s who builds them. Boyer is handing soldering irons and coding tutorials to Native youth like Holden Hoy, turning them from passive learners into active creators of language technology. This approach tackles two crises at once: the accelerating loss of Indigenous languages and the persistent underrepresentation of Native youth in STEM fields.

“

Marie Tribe of Chippewa Indians, has invented SkoBot—a palm-sized AI-powered robot that speaks Anishinaabemowin, an Indigenous language spoken by fewer than 10,000 people in the United States.

The project’s origin is disarmingly simple. Boyer was inspired by Tickle Me Elmo, a toy that teaches through interaction. She realized that no equivalent existed for Anishinaabemowin, a family of languages including Ojibwe, Odawa, and Potawatomi. Her grandmother is one of the last fluent speakers of their dialect, giving Boyer a visceral understanding of how quickly a language can vanish. The SkoBot, which children assemble, wire, and program, becomes a tangible counterforce—one that can speak, listen, and grow with its maker.

Language loss in Indigenous communities is a direct legacy of U.S. Indian boarding schools, which systematically suppressed native tongues for over a century. Today, revitalization efforts often falter because young people may see classroom language lessons as “just another homework assignment,” as Boyer notes. SkoBot reframes the task: instead of memorizing vocabulary lists, students build a physical companion that converses with them. The robot’s AI component—likely using speech recognition and synthesis trained on the language—enables real-time interaction, making language learning immersive and personal.

From an educational technology perspective, SkoBot is a masterclass in engagement. It blends constructionism, culturally sustaining pedagogy, and project-based learning. Students acquire skills in coding, electronics, and AI literacy while deepening their connection to heritage. The end-of-year STEAM Museum showcase where Hoy presented his robot is exactly the kind of authentic assessment that progressive educators advocate for. Moreover, because the robots are built by students for their own communities, the content is authentic and community-vetted, sidestepping the “top-down” tech solution trap that often plagues ed-tech interventions.

The AI angle is equally compelling. Developing natural language processing for a language with only a few thousand speakers and limited digital corpora is a profound technical challenge—and a vital one. Most commercial AI systems prioritize high-resource languages like English and Mandarin. SkoBot represents a growing movement toward low-resource language AI, where community members, not large corporations, drive data collection and model fine-tuning. The project implicitly raises questions about data sovereignty: who owns the recorded speech of elders, and how is it used? Boyer’s model, rooted in community mentorship, offers a template for ethical AI development in cultural contexts.

The implications stretch far beyond Anishinaabemowin. Globally, over 40% of the world’s 7,000 languages are endangered, many with no digital footprint. SkoBot could serve as a replicable open-source blueprint for other communities. Its low-cost, DIY nature—built by kids—makes it scalable without massive funding. The narrative also challenges the tech industry’s obsession with scale and profit, showcasing technology that prioritizes cultural preservation over commercial gain.