On viewFilm & Screen

A.I. Fills the Silence in Documentary on Leonard Peltier

Directors of a documentary on convicted AIM activist Leonard Peltier used A.I. after being unable to interview him — and concede the result "wasn't ideal," used "to mixed effect."

Documentary on Leonard Peltier Uses A.I. to Mixed Effect
Documentary on Leonard Peltier Uses A.I. to Mixed EffectAI-generated

Programme

  1. A new documentary covers the case of convicted American Indian Movement activist Leonard Peltier.
  2. The directors could not interview their subject and used A.I. as their solution.
  3. The filmmakers and coverage acknowledge the approach 'wasn't ideal' and worked 'to mixed effect.'

By Tom Whitfield

The directors of a new documentary about Leonard Peltier, the convicted American Indian Movement activist, used artificial intelligence to construct their film — and the filmmakers themselves concede the solution "wasn't ideal."

That admission, printed plainly in the film's own coverage, is the most honest sentence attached to the project. It deserves scrutiny before the technology gets praised.

What problem was A.I. asked to solve?

The filmmakers faced a hard constraint: they could not interview their subject. Peltier is a convicted man, and the directors had no access to him on camera. Rather than build the film around that absence — absence is, after all, a legitimate documentary subject — they chose to fill it with a machine.

The film concerns the case of the American Indian Movement activist. The case itself has divided opinion for decades. The documentary's innovation is not its argument about the case but its method: A.I. standing in for a voice the directors could not record.

The verdict from the source coverage is measured: the technique was used "to mixed effect." That phrase does a lot of work. It tells us the experiment neither failed outright nor justified itself.

What does this change for documentary filmmaking?

The practical question any viewer or filmmaker should ask: what does this change for nonfiction film? The answer, on this evidence, is — not much yet, and possibly something troubling.

A documentary's core contract is that its words belong to someone who actually spoke them. When a synthetic voice delivers material the subject never recorded, that contract frays. The directors here are at least transparent about the compromise, and the mixed reception suggests audiences and critics noticed the seam.

Note what is not said. The coverage does not claim Peltier approved the A.I. reconstruction. It does not say which parts of the film are synthetic and which are archival. It does not tell us whether the technique was disclosed to viewers inside the film or only in press commentary. Each of these gaps matters, and none is closed.

A caveat for every promise

A.I. is routinely sold as the answer to impossible interviews — the deceased, the imprisoned, the unreachable. This film is an early test of that promise, and by its makers' own account the answer is: not ideal. The technology delivered something; it did not deliver the man.

The honest reading is that this documentary is less a breakthrough than a caution. Future filmmakers facing the same wall now have one case study, and the case study says the tool is blunt.

The Peltier case will continue to draw filmmakers. Whether any of them reach for A.I. again may depend on how audiences weigh the difference between a voice that is accurate and a voice that is real.

Source: NYT Arts

artificial-intelligence, documentary, leonard-peltier, american-indian-movement, filmmaking

Also on view

« Previous articleNext article »