Sam Altman, CEO of OpenAI, mentioned something worth reflecting on during a fireside chat with Tobias Lütke (founder and CEO of Shopify) at Toronto's 2023 Elevate conference:
"If you look at the prediction from maybe 10 years ago maybe even five, I think most experts would have say first AI comes for physical labor. It's going to drive trucks it's going to work in factories. Then it comes from the sort of easier parts of cognitive labor. Then it comes from the stuff that's really hard… Maybe it can write computer code someday - maybe not. And then maybe someday in the distant future (but probably never) it can do creative work. And of course it's gone the exact opposite direction…Almost everybody predicted this wrong."
The quote by Altman summarizes the amazement that many of us experience when interacting with Generative AI, witnessing how it can effortlessly generate high-quality content, whether it's telling a joke in the tone of your favourite comedian or writing a blog post in the style of a famous author.
Not everyone, however, is a fan. Its abilities, though in their infancy, have irked the creative crowd. I am talking about the writers in Hollywood, who are picketing as we speak.
AI: Taking Center Stage within the Scriptwriting Process?
They are worried about the use of artificial intelligence in the movie production process. The writers fear that producers may use AI to write scripts or fill in gaps in unfinished screenplays. This would result in an increased supply of scripts. When the supply of something goes up, the price goes down. That's the way capitalism works. Consequently, the increased use of AI could lead to a decrease in the need for scriptwriters, potentially affecting writers' earnings. While recognizing that this rapidly advancing technology could be a useful tool in some cases, the writers are demanding that production companies agree to certain safeguards regarding its use.
It’s tempting to think that this only applies to those picketing in front of Hollywood. The word “content” may send our minds to art, writing, and movie scripts. However, content generation is not just about the creative crowd. Consider that accountants, lawyers, and consultants are also content generators. Sure, they cater to a different genre and audience. And yes, their content is more likely to make you cry than laugh, it's still content at the end of the day. Moreover, this discourse extends far beyond the confines of movies, television, or streaming platforms, serving as an early warning sign, a proverbial canary in the coal mine, hinting at the impending profound transformations across all sectors and industries.
How good is this technology? Is it really worth protesting?
As they say, seeing is believing. We don’t have to speculate about AI’s script writing powers. We can try it out ourselves. Here is my attempt to do just that. Check it out and see for yourself:
Admittedly, the story was terrible.
It is certainly no competition for the quality and the caliber of those protesting in the Sunshine State. Both my wife and Efrim Boritz thought it needed some work. A lot of work. That being said, if we focus on the aesthetics of the story, however, we miss the point. The story, from start to finish, took a grand total of 35 minutes. Here how I generated the story.
Step 1: Fill-in the Narrative Arc
I took the five elements of Narrative Arc and completed it for my story. This was crafted in the following prompt:
Prompt: Draft the story using the following
“Exposition: A group of auditors working on an audit file for a bank in 2006. There is a deadline of two weeks. Kevin is the hero. He is the senior on the audit
Rising Action: Kevin is asked to audit a mortgage-backed bonds held by the audit client. He tries to understand the structure. He calls the bond rating agency. Doesn't get answers. Tries to call the borrowers no answer. Billy his manager wants him to just get the work done. He wants him to stop asking questions.
Climax: Kevin concludes that the bonds are worthless and must write them down. Billy insists that they use the current trading price. Terry the partner on the file must step-in to address this argument.
Falling Action: Partner decides to go with Kevin. Billy quits and goes to work for Lehman Brothers.
Resolution (Denouement): After the financial crisis, Kevin becomes part of the internal education team teaching him about the importance of skepticism. Billy has lost his job and is having a tough time finding a new one.”
The result was an outline with each element a bit more flushed out. Here’s how “Exposition” looked like:
“In 2006, a group of auditors are working on an audit file for a major bank. The team has a tight deadline of two weeks to complete the task. Kevin, the hero of the story, is the senior auditor on the project. He is responsible for overseeing the audit and ensuring that the team meets their deadline.”
Step 2: Expand out the Story
Then I asked it to take the text and “Expand and write in Michael Lewis's style”. Here is what it came up with:
“In the bustling year of 2006, when the financial world was basking in the warm glow of seemingly endless prosperity, a group of auditors found themselves hunkered down in a fluorescent-lit conference room. Their mission: to dissect and analyze an audit file for a major bank, a task with a ticking clock – a mere two weeks to complete the assignment. The pressure was palpable, the atmosphere charged with a blend of urgency and apprehension.
At the helm of this ragtag ensemble stood Kevin, the protagonist of our tale. A senior auditor with a keen eye for detail and an unwavering sense of duty, Kevin bore the weight of overseeing the audit and steering his team towards the fast-approaching deadline. His steady hand and calm demeanor disguised the internal churn of responsibility and the need to maintain professional skepticism amidst a sea of financial jargon and convoluted transactions.
Kevin's days blurred together as he poured over documents, cross-referencing figures and questioning assumptions. He navigated the labyrinthine world of finance with an uncanny intuition, a skill honed by years of experience that allowed him to cut through the fog of complexity to reveal the hidden truths beneath. Surrounded by his team of auditors, a diverse group of individuals each with their own quirks and strengths, Kevin felt the weight of their collective success or failure resting squarely on his shoulders…”
Rinse and repeat for the remainder of the outline and you’ve got a fully fleshed out story.
We all agree that this is not a work of art, but it only took 35 minutes. Now, imagine what I could create if I spent 35 hours or 35 days on it. Imagine further and think what the studios could achieve with an artificial intelligence that was designed specifically to generate movie scripts.
Was AI-Enabled Scriptwriting Truly Unpredictable?
No, it was not. This sci-fi thriller has been in the making for 20 years. As Chris Steiner describes in "Automate This", Hollywood had access to an algorithm in 2004 that could predict the commercial viability of a script. He writes:
"In 2004, a major movie studio allowed an algorithm to scan nine scripts of unreleased movies. The results of the analysis, run by a new company named Epagogix, were then tucked away. The movies all eventually hit the screen, and when the last one was out of theaters, the movie studio went back to take a look at what the algorithm, which was supposed to tell them how much money each film would gross at the box office, had predicted. In three of the nine cases, the algorithm missed by a wide margin. The other six forecasts, however, were bizarrely accurate. On one movie that the studio expected $100 million or more on, the total gross was $40 million, a huge disappointment.1 The algorithm predicted $49 million. Another prediction was within $1.2 million. Epagogix was suddenly on its way to becoming an indispensable tool for studios to use in analyzing scripts—especially ones that may be attached to big budgets—before the movie gets made. Epagogix was conceived and built by two movie lovers, one of them a lawyer and the other from Wall Street’s favorite of disciplines: risk management. The point is to minimize the risk of producing a stinker like Disney did in 2012 when John Carter lost the studio nearly $200 million."
He then goes on to describe the algorithm that analyzes a script based on a comprehensive report created by humans who evaluate various aspects, such as setting, characters, plot, and moral dilemmas. Despite its advanced capabilities, the algorithm still relies on human judgment to evaluate the script's language, story, and characters. But Steiner asks, presciently:
"What if there were an algorithm that didn’t need people for input? What if there were algorithms that could create the script itself?”
The Deeper Truth about Hollywood’s Existing Algorithmic Approach
There is a deeper truth in what Steiner uncovered: audiences are quite predictable. This notion seemingly contradicts the long-held belief that humans demand an infinite canvas when it comes to creativity, a canvas where one could expect unexpected twists, novel ideas, and a constant reinvention of concepts and narratives. Instead, 20-year-old algorithms are capable of predicting what most people like to watch.
Hollywood is clearly gravitating towards sustaining innovation, instead of pursuing truly disruptive narratives. John Wick is on its fourth instalment. And they just released the tenth instalment of Fast and the Furious. Yes, Fast X. By opting to reiterate tried-and-true storylines, Hollywood ensures its economic prosperity. However, the result is a not-so-creative landscape where sequels are incessantly produced, extending familiar plotlines to an almost infinite degree. This approach provides a measure of security, given the inherent uncertainty of box office returns. Yet it also confines the industry within the bounds of proven narratives, potentially at the expense of groundbreaking, original storytelling.
Hollywood studios are not in the business of searching for avant-garde composers, they're searching for chart-topping artists—those who can consistently produce hits that climb the billboards. And as for box-office bombs, they're a crippling blow to the account books—a $200 million heartache they'd rather avoid.
What about the value of creativity?
What about the pursuit of cinematic excellence, the weaving of a narrative so profound that it moves its audience to tears, laughter, or introspection? That's a narrative they've relegated to the bohemian fringe of society, the ones we affectionately refer to as “starving artists”. That creative crowd can dabble in the intricate arts of filmmaking to their hearts' content, while the Hollywood studios stick to what they know best—churning out billion dollar blockbusters, over and over again
This cold commercial reality births a Faustian bargain—one where artistic vision bows before the altar of profitability. It's the invisible contract that underwrites every script, each casting call, and the red-carpet premieres. It's the unspoken rule, the little secret tucked beneath the glitz and glamor of Hollywood. More on this and Optimus Prime in our next post.
Author: Malik Datardina, CPA, CA, CISA. Malik works at Auvenir as a GRC Strategist that is working to transform the engagement experience for accounting firms and their clients. The opinions expressed here do not necessarily represent UWCISA, UW, Auvenir (or its affiliates), CPA Canada or anyone else. This post was written with the assistance of an AI language model. The model provided suggestions and completions to help me write, but the final content and opinions are my own.
No comments:
Post a Comment