With the immense amount of buzz since its release in June, OpenAI’s GPT-3 has come a long way of deceiving people — starting from creating a fake blog to writing opinionated articles along with posting Reddit comments and roasting Elon Musk’s tweets.
Such advance tasks handled by GPT-3 made people, as well as researchers, realise its immense potential of creating artificial general intelligence. The model not only learned how to code but also to compose music, art, poetry as well as do mathematics — been applied to many interesting ways. Adding to its accomplishments, GPT-3 has now come up with a short film screenplay — Solicitors.
An approximately 4 minutes short film — Solicitors — was written by the GPT-3, which isn’t the best screenplay but is even not the worst, considering a machine has written it. The script was initiated by a few lines, written by two of senior student filmmakers from Chapman University, that was fed on to the machine, and the rest of the screenplay has been generated by leveraging the massive language model.
Subscribe to our Newsletter
Join our editors every weekday evening as they steer you through the most significant news of the day, introduce you to fresh perspectives, and provide unexpected moments of joy
The film starts with a man knocking on a woman’s door to share a story about an accident he has been involved in. With the machine writing the screenplay, it has become a bit difficult for the audience to understand it, but it claimed to have a twist ending.
GPT-3 Learns To Make Films
To write the AI-generated screenplay, the students used a GPT-3 tool called — Shortly Read, which documented the entire screenplay post the first 20seconds. The lines that were fed on to the system were — “Barb’s reading a book. A knock on the door. She stands and opens it. Rudy, goofy-looking, stands on the other side.
With GPT-3’s 175 billion parameters that it has been trained on works to bring out the associations. The algorithm draws a parallel between words and phrases based on its training data. According to the news media, the story seems to be a little odd and also not so compelling or character-driven but considering its written by machines; it indeed got a benefit of the doubt.
As a matter of fact, the GPT-3 tool — Shortly Read — has a tagline of “Never experience writer’s block, ever again,” which has been designed to inspire writers and authors with their content when they face any sort of blockage while writing stories. Although this sounds good, many experts believe that “how much help can a writer get from a tool like this before it becomes something like cheating?”
Well, this isn’t the first time where an AI tool has been used to write a screenplay or make a movie; however, this might be the first time, AI tool has been specifically trained to write screenplays. And, according to experts, now that GPT-3 is capable of writing scripts, what else it can be good at — legal briefs, news articles, political analyses, letters to your relatives, etc. It is the advancement of the tool and its potential in a plethora of use cases that made the algorithm’s creators be in a dilemma of releasing it altogether.
When asked, media futurist and algorithmic filmmaker Alexis Kirke, mentioned that it is crucial for us to get used to the idea of artificial intelligence having a hand in our creative endeavours, especially when it comes to writing movie scripts.
He stated to the media — “A huge amount of experience has been codified by writers, producers, directors, script editors, and so forth.” However, currently, if anybody wants to reduce the number of adverbs and adjectives in their script, there is an algorithm for that. If anybody wishes to ensure their characters’ dialogue all sound[s] different from each other, there is an algorithm for that. There is even an algorithm to generate alternative, less clichéd, rewrites of a page that keep its general meaning. Thus we all should get used to this fact. With algorithms being there for almost every task, GPT-3 has made its position to be the most accurate among those.