This is an AI-generated image created with Midjourney by Molly-Anna MaQuirl
When it comes to discussing the world of film and television, AI news tends to be fairly negative. With the world watching the writers’ and actors’ strikes, a lot of that discussion has focused on whether writers can be replaced by AI.
One of the primary negotiating points of the SAG-AFTRA actors in the Netflix strike is for protections from having their likeness used by AI without their consent or adequate compensation. The Writers Guild of America strike is fighting a similar battle against AI being used to write scripts. These are both fairly unsettling prospects and it’s easy to see why so many in the entertainment industry have chosen to take a stand.
AI is cheaper and faster than traditional creative methods. When you only need to pay a single AI manager instead of an entire writers’ room or only pay an actor once to have their voice and face captured for an AI database, it seems like a smart business decision. In the short term, AI taking center stage could be a real possibility in the industry.
One thing that many people who are excited about the idea of AI taking over film and television writing don’t entirely understand is that AI isn’t actually capable of creativity — at least not the AI we have today. AI works by assimilating data and then breaking down what it has processed into a new form. The material is technically new, but it isn’t truly a creative product.
For example, you could download every movie that Alfred Hitchcock made into an AI and give it the task of creating a movie. The movie it created would be suspenseful and full of twists with a classy vintage aesthetic. However, the entire thing would simply be elements of Hitchcock’s works adapted into a new form.
Where AI is more practical is in voice acting. If you’ve ever wondered “Will AI replace voice actors?”, –the answer is not entirely. Voice actors will still be needed to provide the raw material for AI to work with. There are already consumer apps on the market that use small snippets of recorded speech to fully replicate an individual’s speech patterns and voice. This could easily be expanded to allow AI to replace voice actors in animations or to dub foreign language films.
What most actors are concerned about is generative AI. Generative AI comes in many forms. It’s the blanket term used to describe any AI that is used to generate images, text or any other type of media. These AI solutions can range from incredibly basic to incredibly complex. ChatGPT is a text version of generative AI that has been getting a lot of attention recently and Dall-E is a generative AI for images.
The form of generative AI that actors are concerned about is a more advanced tool. It is an expansion of the deepfake technology and is already being used to make actors look younger. By capturing a full range of an actor’s expressions, movements and vocal patterns, the AI can produce a digital copy that is eerily convincing.
We’re still at a point where this generative AI is used more as filter over an actor’s face or body. The worry is that soon AI will be able to do all the work itself and the actor won’t have to be present at all. This may take a few years still, but it’s the reason that actors are being so proactive and trying to have protections established now.
It isn’t exactly the AI itself that’s a problem. The bigger issue is about consent and compensation. Many actors have stated that they would be fine with the option to ‘rent’ or license their image to an AI studio under controlled terms, such as having a say in what projects their likeness was used in and being paid for it.
This makes a lot of sense. Actors get to choose their roles normally, so why shouldn’t they get to choose the roles that their AI version plays? As for compensation, it would make sense for it to look similar to any other type of media licensing agreement. Studios are required to pay a fee for the songs that they use in the soundtrack, and this would be a similar process.
If studios and actors don’t manage to come to an agreement, we might be entering a new phase of copyright law. If the only way for actors to protect their likeness and name is to copyright it, this will become the norm. There’s currently a bit of a grey area surrounding a person’s ability to copyright their own likeness, but that just might change rapidly.
It seems likely that AI in film and television will follow a similar pattern to CGI. In the first years of CGI special effects, everyone thought they were amazing and would completely replace practical effects. For several years, it looked as if that was going to be the case.
Slowly, however, directors and audiences began to see the value in practical effects again and realize that they make a film so much more visceral and impressive. While CGI will always have a place in the industry, practical effects are more respected and valued. You only need to look at a film like Christopher Nolan’s Oppenheimer (2023), which only used practical effects, to see just how impressive they can be and how they can never be fully replicated by a computer.
Once the appeal of novelty has worn off, the desire for scripts written by real, talented writers will return. As intelligent as computers can be, the ‘artificial’ is in the name for a reason and they can only replicate and refashion, not create independently.
When it comes to AI taking center stage, whether the big studios will remember the value of writers in time remains to be seen, but we can simply hope that we aren’t watching the death of a creative industry.
This is an AI-generated image created with Midjourney by Molly-Anna MaQuirl
When it comes to discussing the world of film and television, AI news tends to be fairly negative. With the world watching the writers’ and actors’ strikes, a lot of that discussion has focused on whether writers can be replaced by AI.
One of the primary negotiating points of the SAG-AFTRA actors in the Netflix strike is for protections from having their likeness used by AI without their consent or adequate compensation. The Writers Guild of America strike is fighting a similar battle against AI being used to write scripts. These are both fairly unsettling prospects and it’s easy to see why so many in the entertainment industry have chosen to take a stand.
AI is cheaper and faster than traditional creative methods. When you only need to pay a single AI manager instead of an entire writers’ room or only pay an actor once to have their voice and face captured for an AI database, it seems like a smart business decision. In the short term, AI taking center stage could be a real possibility in the industry.
One thing that many people who are excited about the idea of AI taking over film and television writing don’t entirely understand is that AI isn’t actually capable of creativity — at least not the AI we have today. AI works by assimilating data and then breaking down what it has processed into a new form. The material is technically new, but it isn’t truly a creative product.
For example, you could download every movie that Alfred Hitchcock made into an AI and give it the task of creating a movie. The movie it created would be suspenseful and full of twists with a classy vintage aesthetic. However, the entire thing would simply be elements of Hitchcock’s works adapted into a new form.
Where AI is more practical is in voice acting. If you’ve ever wondered “Will AI replace voice actors?”, –the answer is not entirely. Voice actors will still be needed to provide the raw material for AI to work with. There are already consumer apps on the market that use small snippets of recorded speech to fully replicate an individual’s speech patterns and voice. This could easily be expanded to allow AI to replace voice actors in animations or to dub foreign language films.
What most actors are concerned about is generative AI. Generative AI comes in many forms. It’s the blanket term used to describe any AI that is used to generate images, text or any other type of media. These AI solutions can range from incredibly basic to incredibly complex. ChatGPT is a text version of generative AI that has been getting a lot of attention recently and Dall-E is a generative AI for images.
The form of generative AI that actors are concerned about is a more advanced tool. It is an expansion of the deepfake technology and is already being used to make actors look younger. By capturing a full range of an actor’s expressions, movements and vocal patterns, the AI can produce a digital copy that is eerily convincing.
We’re still at a point where this generative AI is used more as filter over an actor’s face or body. The worry is that soon AI will be able to do all the work itself and the actor won’t have to be present at all. This may take a few years still, but it’s the reason that actors are being so proactive and trying to have protections established now.
It isn’t exactly the AI itself that’s a problem. The bigger issue is about consent and compensation. Many actors have stated that they would be fine with the option to ‘rent’ or license their image to an AI studio under controlled terms, such as having a say in what projects their likeness was used in and being paid for it.
This makes a lot of sense. Actors get to choose their roles normally, so why shouldn’t they get to choose the roles that their AI version plays? As for compensation, it would make sense for it to look similar to any other type of media licensing agreement. Studios are required to pay a fee for the songs that they use in the soundtrack, and this would be a similar process.
If studios and actors don’t manage to come to an agreement, we might be entering a new phase of copyright law. If the only way for actors to protect their likeness and name is to copyright it, this will become the norm. There’s currently a bit of a grey area surrounding a person’s ability to copyright their own likeness, but that just might change rapidly.
It seems likely that AI in film and television will follow a similar pattern to CGI. In the first years of CGI special effects, everyone thought they were amazing and would completely replace practical effects. For several years, it looked as if that was going to be the case.
Slowly, however, directors and audiences began to see the value in practical effects again and realize that they make a film so much more visceral and impressive. While CGI will always have a place in the industry, practical effects are more respected and valued. You only need to look at a film like Christopher Nolan’s Oppenheimer (2023), which only used practical effects, to see just how impressive they can be and how they can never be fully replicated by a computer.
Once the appeal of novelty has worn off, the desire for scripts written by real, talented writers will return. As intelligent as computers can be, the ‘artificial’ is in the name for a reason and they can only replicate and refashion, not create independently.
When it comes to AI taking center stage, whether the big studios will remember the value of writers in time remains to be seen, but we can simply hope that we aren’t watching the death of a creative industry.
Navigating the World of Artificial Intelligence
Copyright 2024 AI Atlas. All Rights Reserved. Registered address: Union Building, Rose Lane Business Centre, 51-59 Rose Ln, Norwich NR1 1BY. UK.
Search
Menu