Prime Video announced on Thursday that it is adding AI lip sync to its dubs: software that changes the movements of actors' mouths so they match the English dialogue rather than the language the performance was recorded in. It starts with one title, the English version of the German series Maxton Hall, and Amazon says it intends to bring the feature to other projects. The company has not released a single frame showing the result. For a visual effect applied to human faces, that absence is most of the story.
Amazon has not said how much of the work the AI automates, or how the feature will actually be deployed. Raf Soltanovich, vice president of technology at Prime Video and Amazon MGM Studios, said the company is constantly building tools to improve viewing, and that lip sync should make discovering content from around the world smoother and more immersive. The New York Times reported the announcement.
Everything else about the launch is open. Whether the feature will be available across all films and series. Whether rights holders get any say in its use. Whether an individual actor's distinctive facial movement survives the process or flattens into generic mouth motion. Whether the output is convincing at all, or lands in the range of strange facial distortion that this class of effect has historically produced.
This is not Amazon's first pass at the problem. Last year Prime Video launched an experimental "AI-assisted" dubbing feature for a dozen films and shows in English and Spanish. The company has also tested AI-generated recaps of television series and a tool that surfaced recommendations based on similar storylines and character arcs. Taken together, the direction is consistent: Amazon is trying to remove every visible seam from the experience of watching its own catalogue.
A one-title launch with no footage reads less like a product release than a temperature check. Announce it, let the trade coverage happen, see whether anyone objects loudly enough to matter, then decide how wide to go. The absence of a demo is doing work here too, because a lip sync effect is judged entirely on how it looks and Amazon has chosen to be judged on how it sounds described. Companies with a result they are confident in usually lead with the clip.
The more interesting question is what the feature is for. Dubbing has real justifications, particularly for viewers with disabilities, and nobody needs to defend the existence of a dubbed track. But dubbing has always announced itself. The mismatch between the mouth and the sound is the viewer's reminder that a translation is happening, that the performance underneath belongs to someone speaking another language. Lip sync removes that reminder on purpose. The actor's voice was already replaced; now the face is edited to hide the replacement.
What Amazon is proposing to smooth away is not an inconvenience. It is the last piece of evidence in the frame that the thing you are watching was made somewhere else, by people who were not speaking English.