By Dr. Robert E Furey
A few decades ago, I was teaching a course covering science and technology in society. I wanted to show students a world where anything will be possible, both marvelous and mundane. We looked at knowledge and applications, and just how they shaped the quotidian aspects of everyday life. Things you might not readily think of like concrete or lightbulbs, antibiotics, or the postal system. In some cases, images of society before and after or with and without even the most mundane technology revealed some interesting insights into how society works.
Once we had familiarized ourselves with the generalities of how society adopts change, we could play around with the concepts and predict the directions of technology and its impacts on society. Because, as should be obvious, the two are as tightly intertwined and complementary strands of DNA. When one side moves, the other catches up only to move a little farther along, then the first side responds in kind in a never-ending, self-referring cycle of socio-scientific modification.
I always found this capstone to the class enjoyable and insightful. Most of the changes students brought up were along the lines of flying cars or affordable moon tourism and never things like more efficient bottle caps.
One of the more controversial ideas I introduced, controversial in that the majority of participants found it utterly preposterous, had to do with the future of cinema. I made this proposal over two decades ago, but it seemed as obvious to me then as it does today. And in the meantime, quite a few of the necessary pieces have been put into play that are clearly harbingers of its deployment.
So, what was the application?
What could embody the “Magic of Hollywood” more than a movie starring actors you thought were gone from the world? Imagine an action flick where Bruce Lee finds himself facing the evil mechanizations of Vincent Price. Or a dramatic comedy crossover starring Steve McQueen teaming up with Laurel and Hardy to save a young Judy Garland? Any actors with any other actors doing absolutely anything.
Let’s back up a little here and follow the stepwise progress to digital movie stars.
In 1950 Alan Turing described the Imitation Game. Participants would engage in natural language conversations through a text-only interface with unseen partners, some human and some machine. Since then, the Imitation Game has become the Turing Test and the machine participants are now Chatbots.
Chatbots and AI have coupled in some remarkable ways that would have beaten the original Turning Test conditions. Annual Turing Tests have become more difficult and moved the bar enough that even the best chatbots fail. Although they have moved into speech-abled bots that I am sure we have all talked with a few times on the phone and not even known. You can trip them up if you suspect they’re there, but for most of us and most conversations, it’s hardly worth the effort.
Lately, Chatbots have become so sophisticated and trainable that people are using them to relive the loss of a loved one. On the one hand, people preparing for their own demise sit for hours of video and audio recording that AI can reformat into a simulacrum of that person, a sort of holographic chatbot that can interact with friends and family long after you’re gone. If you are interested in this for yourself, look into StoryFile. A visit to their website offers an interactive conversation with William Shatner.
For a second arrangement along the same idea, you feed data into a chatbot for training; memories, emails, mail, photographs, text messages, everything you have, and out pops an interactive personality with the knowledge of the deceased. Unlike the first example, this chatbot does not require the participation of the modeled person, just what they left behind. Microsoft holds a patent on this technology. They report that the results are so disconcerting that they will not release the technology.
Both of these methods produce interactive entities built on real data from real people. We can think of them as a skeletal framework upon which to affix complementary technology. But what do we need for autonomous actors? We need appearance and action.
Deepfake technology has been available for some time now. Deepfake will splice the likeness of anyone’s face onto the face of another. It has gotten very good, so good in fact that compromised videos making their way into our news streams is an important issue. So bad for truth but good for entertainment. Recently, Bruce Willis announced he had sold his likeness to the Russian production company Deepcake to continue making movies he can no longer act in. While his people have denied the report, the fact remains that he could do it if he chose to.
In a similar story, James Earl Jones sold the rights to his voice. An AI has been trained on his voice and will continue to perform as the voice of Mr. Jones. These efforts are moving forward with a technology called Respeecher, an AI able to listen to a voice and seamlessly repurpose the sounds into new words and phrases. This is not a voice changer; this is a new AI conversation.
Once physical and auditory appearances are covered, motion comes to play. We’ve come a long way from Ray Harryhausen’s “dynamation.” Stop motion animation has been replaced with ultra-sophisticated CGI where anything is possible. It can be expensive, but as computer power increases, rendering costs come down. One has only to watch the spectacularly forgettable movie “Avatar” to see some of what visual magic can be done.
We have really slick chatbots that can learn and assume the personalities of anyone with a history. We have actors selling their voices, and actors selling their likenesses, both done specifically to continue an acting career coming to a close through age or illness. We have the ability to render any environment and move through in any manner we choose.
None of these technologies were available 20 years ago.
It is but a small leap to imagine actorbots. To make our dramatic comedy we set one to watch Steve McQueen in everything he ever did captured on celluloid. Every movie, every talk show, record every nuanced motion in his movements and expressions, the inflections of his voice. Soon enough our AI goes by the name of Steve. Do the same for Stan, Oliver, and Judy. We’re all set.
We’ll have no need for cameras and film crews, only AI technicians and babysitters. No more physical sets of course. Production costs plummet. People will warn of stagnation in Hollywood now steeped in the movie stars of yesteryear being constantly flogged into new scenarios. While that may occur, it won’t last long.
This will mean the democratization of movies. Anyone with a big enough computer and access to actorbots will be making movies. Since actorbots are trained, shuffling and tweaking the personalities will be easy enough that we can expect specific players for each story. There will be no need for movie stars. And maybe it was the no-movie-stars-needed that caused all the pushback.
But pushback has not stopped the advancement.
Tilly Norwood, the first AI actress, was born in September 2025. Her arrival brought fear, denial, and threats of boycotts to any company that might try to represent her. Amid the international backlash, Tilly began appearing in short clips on talk shows, mock movie previews, and finally her own music video calling direct attention to her electronic pedigree.
This year the Cannes Film Festival included the first entirely AI generated full length film. “Hell Grind” is a standard 95-minute movie produced in just two weeks. A crew of 15, with a budget of only about $500,000, completed the project.
Of course, SAG (Screen Actors Guild) is not happy. The WGA (Writers Guild of America) is in panic mode. Organizations in the UK and Canada are in lock step. Hollywood has responded with a list of things to make a film ineligible for awards, AI being high on that list.
The thing is, AI is here to stay now. Hollywood may be busy piling sandbags and shouting about the grave danger of AI encroachment, but the coming effects will be far wider spread.
When something as complex as producing a movie, often done with budgets in the 100 millions, can be done with bankroll in the 100 thousands, bulwarks against AI seem futile.
We can talk about AI shortcomings, hallucinations, the lack of humanity (such as it is), and glitches, the fact is, it’s here. AI will encroach on and change almost every aspect of our lives.
But I am getting ahead of myself here. We will have electronically resurrected actors in new roles in new movies. The next time Bruce Willis saves the planet, Buster Crabb will be there to help. Or maybe Bruce will be there to help Buster. After all, anything goes. Tilly will see to it.
Dr. Rob Furey is an evolutionary biologist, behavioral ecologist, and forensic scientist. He has been teaching integrated sciences and technology classes for 30 years. He has also been a forensic entomologist for 20 of those years. Currently, he is Chair of Forensic Sciences at St. Edward’s University and sits on the university’s Artificial Intelligence committee.