WRITTEN BY… / PART 21: THE WRITER VS THE MACHINE
Alex Proyas - THE FILMMAKERS SURVIVAL GUIDE
No discussion about the future of storytelling can avoid the subject that now hangs over every creative discipline: the machine.
Artificial intelligence has arrived in the arts with startling speed. Tools capable of generating text, images, music, animation, and even entire sequences of film are appearing almost monthly. Some greet this development with excitement. Others with dread. Many with a confusing mixture of both.
For writers and filmmakers, the question is unavoidable: what happens when machines can create stories?
To answer that properly, we have to separate two ideas that are often blurred together in public debate. One is automation. The other is authorship.
Automation has always been part of filmmaking. Every technological advance in cinema - from sound recording to editing systems to digital effects - has automated something that previously required enormous labour. Previously filmmakers physically cut and glued strips of film together – I still remember how frustrating it was losing film trims in the cutting room. Digital editing replaced that process with software timelines and efficiency. Visual effects that once required miniatures and optical printing can now be generated far more easily in a computer.
Each technological shift removed a layer of mechanical difficulty. But it did not remove the need for imagination.
What artificial intelligence introduces is a different kind of tool. Instead of automating physical labour, it automates certain forms of creative labour. It can suggest dialogue. Generate story outlines. Produce concept art. Simulate voices. Assemble images into motion. In some cases it can even mimic the style of existing filmmakers or writers. None of it necessarily better than a human could do, though certainly faster.
This understandably makes artists uneasy. I confess I was at first. If a machine can eventually generate an entire screenplay or produce a whole movie, where does that leave the human storyteller?
The answer depends on how we define storytelling in the first place.
A machine can generate patterns. It can recombine existing information. It can predict what kinds of narrative structures are statistically likely to appear in successful stories. But what it cannot do is experience the world. It cannot suffer loss, feel love, struggle with moral conflict, or wrestle with existential questions. It cannot live inside the chaos of being human.
Storytelling, at its deepest level, emerges from that chaos.
Every story worth telling begins with a question about the human condition. Why do people behave the way they do? What does it mean to be free? What happens when identity collapses? Can we survive lost love? These are not computational problems. They are human ones.
Artificial intelligence can certainly help shape the vessel that carries a story, but it cannot generate the emotional gravity that gives a story meaning. It may prove useful as a script editor, a researcher, even a diagnostic tool for structure. But ask it to write an entire screenplay and it falls on its face. I know, because I’ve tried it out of curiosity.
That distinction matters. The moment you treat AI as a replacement for authorship, the work begins to die. What it produces may be structurally serviceable, occasionally even slick, but it lacks the strange volatility that comes from human experience. It can imitate the appearance of storytelling, but not the pulse of life behind it.
But if you treat AI as a tool - one among many - the picture changes.
The history of cinema is the history of artists adopting new tools and bending them toward creative ends. The camera itself was once considered a machine that threatened traditional art forms. Editing was once seen as a mechanical intrusion into performance. Digital cinematography was once dismissed as inferior to film - until the quality improved beyond what chemistry was capable of reproducing.
In each case, artists eventually discovered that the new technology expanded the language of cinema rather than destroying it.
AI may prove to be another such expansion.
Imagine a filmmaker who can rapidly visualise worlds that would previously have required enormous budgets. A writer who can generate exploratory visual material while developing a script. A small creative team capable of producing imagery that once demanded the resources of an entire studio.
This is where the concept of hybrid filmmaking becomes interesting.
Hybrid filmmaking is not about surrendering creative control to algorithms. It is about combining human authorship with machine-assisted production. The writer conceives the story. The filmmaker shapes the vision. The machine accelerates certain processes that once limited what could be attempted.
In this model, AI functions almost like a form of cinematic augmentation. It extends the reach of the storyteller.
But the storyteller must remain the centre.
There is also a danger worth acknowledging. Machines trained on vast archives of existing media inevitably reproduce patterns from the past. They excel at imitation. Left unchecked, this can produce an endless stream of derivative content - stories that feel familiar but empty, constructed entirely from recycled fragments of previous work. If creators rely too heavily on these systems, we will see a cultural flood of competent but soulless storytelling. Perfectly structured scripts with nothing urgent at their core. Visually impressive films that contain no real point of view.
The responsibility therefore falls back on the writer.
Your job is not simply to generate narrative material. Your job is to bring meaning into the work. To inject the unpredictable, irrational, emotional complexity that machines cannot replicate.
For those outside the arts, there is perhaps one crucial thing to understand, something most artists take for granted: we actually enjoy the process of creation. We do not merely endure the labour in order to arrive at the finished product. The struggle is part of the reward. The wrestling with an idea, the false starts, the problem-solving, the long hours of doubt and discovery - this is where much of the meaning lies for the artist. We do not want to hand that over to machines, because in doing so we would surrender the very thing that makes the work fulfilling. The satisfaction comes not simply from having something finished, but from having made something that could only have come from us, however imperfectly, through effort, instinct, and will.
In other words, the writer becomes more important, not less.
The tools may change. The production methods may evolve. But the fundamental need for a human consciousness behind the work remains unchanged.
In fact, as machines become capable of producing increasingly convincing simulations of storytelling, audiences may begin to crave something unmistakably human.
Something idiosyncratic. Something flawed and personal. Authenticity becomes a kind of currency.
This brings us back to a theme that has run through this entire series: authorship.
The writer who understands their own voice - their obsessions, their questions, their philosophical concerns - will always have an advantage over a system that merely recombines existing material. Machines can reproduce style. They cannot generate a worldview.
And cinema, ultimately, is the expression of a worldview.
We are entering a period where the tools of filmmaking will become more powerful and accessible than ever before. At the same time, the temptation to outsource creativity to machines will grow stronger. Endless slop will be churned out – it already has.
The challenge for the next generation of storytellers will be to use these tools without surrendering to them.
To harness automation while preserving authorship. To build hybrid forms of cinema where technology expands imagination rather than replacing it.
But the story still belongs to the human writer.
