Access to communication is a fundamental human right which, in cascading crisis situations, is often inhibited by language barriers and lack of cultural and contextual understanding.

For communities at risk, simultaneous access to multiple languages and across multiple spoken, written, and visual modes is essential. As recent earthquakes, floods, fires, terrorist shootings, and COVID-19 lockdowns have demonstrated, Aotearoa New Zealand is unprepared for timely crisis communication across the over 160 languages and cultures of its communities.

Although artificial intelligence can be used to augment human translation, does it fulfil its promise of accuracy, respect, and cultural sensitivity in the face of real-world, time-critical applications?

Taking an innovative, complex systems approach that balances computational and humanistic expertise, we investigate how Human-AI collaborative intelligence in the context of crisis communication fundamentally reshapes translation practices, informs ethical questions of authorship, agency, and appropriation, and contributes to sustainable economic, social and cultural transformation.