When you see the face of a familiar actor on screen, you probably assume that somewhere, some time, the actual human being presented to you via technology was really in a studio in front of a camera, speaking the lines you hear. It is only when we remember that motion pictures are designed to produce illusions that we realize the words we hear may be another actor’s voice-over, the background may be green-screened in, and even the actor’s face could have been digitally retouched, or even created from scratch with sophisticated software.
Thinking about these things distracts from the enjoyment of the movie, so usually we don’t. But if you saw a person looking, moving, and sounding exactly like Humphrey Bogart acting in a movie made in 2015, say, it would be hard to ignore the little detail that Bogart died in 1957.
The 1994 film “Forrest Gump” digitally placed the live actor Tom Hanks in archival footage of famous deceased persons such as John F. Kennedy, but what I’m talking about is the reverse: hauling John F. Kennedy out of the grave to make him play a role in, say, a new Judd Apatow comedy. And here’s where we get into some ethical qualms.
In a recent New Yorker article, the digital exploits of University of Southern California computer scientist Paul Debevec are described, from his early work reverse-aging Brad Pitt in “The Curious Case of Benjamin Button” (2008) to his current attempts to preserve holographic versions of Holocaust survivors for use in a permanent museum display. If the latter project is successful, visitors will be able to pose questions and a three-dimensional representation of the person, accurate down to such details as shadows that fall naturally according to the room lighting prevailing at the time, will answer the questions via artificial-intelligence technology. In effect, Debevec will have resurrected the dead, although to a strictly circumscribed sort of life.
For the last few decades, progress in digitally-enabled technologies that initially depend on huge amounts of computer processing have followed a consistent path. First, a new technique is developed at great expense, often paid for by the military or government agencies, and demonstrated in a limited way. Next comes commercialization, with large institutions and corporations being first in line to use it. And finally, advances in hardware and software lower the cost enough to make it affordable to a reasonably large number of average citizens. Debevec fully intends for his super-accurate simulation and illumination technology to follow this well-worn path, so we ought to give at least a little consideration to its ethical implications.
The fact that Debevec is labeled a scientist obscures the reality of what he is doing when he takes Angelina Jolie’s picture from hundreds of different angles to make a digital clone of her to perform a film stunt too dangerous for stunt doubles to do: he is being an artist. And the ethical rules for artists doing art are different from the rules for scientists doing science. From what little I know about the way art is regarded in Western cultures today, there aren’t any ethical rules that are generally observed, unless you count legal strictures such as bans on child pornography and copyright laws. I suspect if Judd Apatow tried to make a digital clone of John F. Kennedy to do the kind of disreputable things that actors in his films typically do, he might hear from the Kennedy estate via a process server. But the legal treatment of public figures differs from the way the rights of a private citizen are treated. Courts have held that as long as the image of a public figure is not being used for commercial exploitation, it is okay to portray it in a work of art. After death, a person’s estate can still control the use of the person’s image for commercial purposes, and this would undoubtedly include a Debevec-style holographic image. So when Debevec was asked if he had considered resurrecting Marilyn Monroe, for instance, he said that Monroe’s estate was unwilling, and so he dropped the idea.
On the other hand, when ConAgra, the firm that makes Orville Redenbacher popcorn, approached Debevec to simulate the recently deceased popcorn king for a TV commercial, Debevec readily agreed. So, two years after Redenbacher died in 2005, viewers saw a digital version of Redenbacher, still promoting his popcorn. But when critics started referring to “Orville Deadenbacher, the popcorn zombie” the ad disappeared. This was not a violation of ethics so much as it was a violation of good taste, and I’m not talking about popcorn.
Highly realistic holographic images of people, alive or dead, are simply the latest advance in a sequence that began 40,000 years ago, when someone blew paint through a stencil onto the wall of the Cave of El Castillo in northern Spain to form the earliest known cave paintings. Sculpture, portraits in oils, photography, motion pictures, and CGI (computer-generated images) followed, and it is only our inordinate addiction to novelty that makes us think there is something fundamentally different in Debevec’s hyper-realistic representations of the human form. Art is one of the most notable activities that separates humans—the only rational animal—from other animals, and the fact that Debevec’s form of art involves rationality of a scientific and technological kind does not make it any less an art form. And art can be put to both the holiest and the most debauched of uses.
As our power to create increasingly realistic-looking digital human forms grows and the technology to do this spreads, we can only hope that artists will rediscover that truth, beauty, and goodness are their true subjects. Most of the art world refuses to acknowledge this fact, which is the real basis for the ethics of art. But no amount of technological advance will change that situation. That change cannot take place in server rooms, or in the theories of computer scientists. That change can take place only in the heart.
Karl D. Stephan is a professor of electrical engineering at Texas State University in San Marcos, Texas. This article has been republished, with permission, from his blog, Engineering Ethics.