Protect Foreign Language Learning from the AI’s Universal Translator
AI is making big advances that can degrade our cognitive potential in ways we don’t realize
People have always wanted to be able to talk to each other easily across languages and cultures.
As a child, I recall watching Star Trek, in which Captain Kirk operated a device about the size of a modern cell phone that translated the languages of different alien species encountered by Earth Federation.
With the recent progress in artificial intelligence (AI), that goal seems closer than ever.
AI-powered translation tools like those made by Spotify, Meta and AI startups promise to break down language barriers and make it easier for people from different cultures to talk to each other. Underneath this optimistic view of technology, however, there are some scary problems that could happen.
Even how exciting this technology brings to many fields of commerce, culture, science and art, we have to consider the tradeoffs as well.
Concerns about new AI translation technologies include losing nuances, reinforcing biases, putting jobs at risk, and even making it possible for new kinds of information warfare. Should we make AI that can translate between languages better than humans just because we can now do it? And if so, how can we use these technologies that could change society in a responsible way?
Will universal translators that use AI bring people together or get lost in translation? We also need to think about what they mean for people and society.
After all, there is a lot more to language than just words. The ways in which we exchange information have profound effects on our social interactions and worldviews. Without careful implementation, these technologies threaten to upend vital social and cultural interactions.
There is a lot to consider and too many issues to cover in just one post. So, I want to focus on the cognitive, educational and epistemological risks of adopting the technology.