A new wave of artificial intelligence tools is emerging as a potential lifeline for endangered and Indigenous languages around the world, sparking optimism among linguists, cultural preservationists, and communities fighting to keep their native tongues alive.
Once seen as a force primarily shaping the future, AI is now being embraced as a powerful tool to protect the past particularly by helping document, teach, and revitalize languages that are at risk of disappearing within a generation.
Open source models, voice recognition tools, and machine translation software are being rapidly adapted to support minority languages. Tech companies and research institutions are collaborating with Indigenous groups and language experts to create digital archives, AI-powered learning apps, and conversational agents that can understand and generate speech in languages with few living speakers.
“Language is more than just communication it’s identity, culture, and history,” says Dr. Leilani Harper, a linguist working with First Nations communities in Canada. “The potential for AI to support revitalization efforts is unprecedented, especially for languages that have limited written records or formal teaching materials.”
One standout initiative includes the development of neural language models trained on oral histories and archival recordings. These models can reconstruct grammar, syntax, and vocabulary patterns, allowing researchers to simulate fluent speech even in languages where only a handful of speakers remain. In some cases, AI has been used to rebuild extinct languages from fragmented texts, enabling educational tools for younger generations.
Earlier this year, an AI research team in New Zealand collaborated with Māori elders to build a voice assistant capable of understanding and responding in te reo Māori. Similar projects are underway in parts of Africa, Southeast Asia, and the Arctic, where linguistic diversity is rich but often overlooked in global tech development.
A key breakthrough has been the ability of modern AI models to learn from limited data — a significant advantage for languages with few speakers or scarce documentation. Unlike older systems that required massive datasets, today’s models can be fine-tuned with just a few hours of spoken audio or written text.
However, the use of AI in language preservation is not without ethical considerations. Critics warn that tech companies must work closely with communities to ensure that cultural knowledge is treated with respect and that AI tools do not appropriate or commercialize sacred traditions.
“There’s a real danger in treating languages as just another dataset,” cautions Professor Samuel Okonkwo, a digital humanities scholar at the University of Lagos. “What matters is empowering communities to control how their languages are represented and used.”
To address these concerns, some projects are being built with open governance models, allowing Indigenous leaders to determine how data is collected, stored, and shared. Others are using blockchain to track digital ownership and ensure that communities retain intellectual rights over their linguistic heritage.
As global awareness of linguistic loss grows with UNESCO warning that nearly half of the world’s 7,000 languages could vanish by the end of the century the use of AI offers a rare blend of innovation and urgency. Far from replacing human voices, these technologies may help amplify them, preserving centuries of wisdom and identity for generations to come.
In an age of accelerating change, it seems that some of the most powerful technologies of the future may play a crucial role in safeguarding the voices of the past.
source: edition.cnn.com