AI has begun to reshape literary translation, not only in how texts are rendered but in how we understand language, meaning, and interpretation.
Over the past several years, I’ve explored this problem through talks, teaching, and a series of experiments and reflections collected on this site. The results are not conclusive. But certain patterns—and certain limits—are becoming clear.
Some insist that literary translation is fundamentally resistant to automation, that machines will continue to make not just ordinary mistakes, but errors that are qualitatively different from human ones. They are right.
Others point to new possibilities, claiming that with the right prompts and methods, AI can open approaches to translation that were previously unavailable. They are also right.
What follows is a set of entry points into the tensions and sometime anxieties of AI and translation.
Experiments with AI Translation:
Core Problems in Translation: More Broadly:These posts are not a system but an ongoing attempt to understand a problem that is still taking shape. I’ll continue to add to them as the technology—and our understanding of it—develops.
###
Russell Scott Valentino 