Filmmaking / Directing : Will you embrace AI as part of the future of filmmaking? by Vincent Weberink

Vincent Weberink An ongoing discussion in the IPG for the last several years. At Diamond Shadow Productions, we use AI for what it is good for, throughout the production and post process.

There are invaluable AI tools developed over many years, of use to the professional industry. Some of the types of applications you list here are aimed at, and marketed to, a consumer level filmmaker. They are not really capable of adding value to the professional production process.

We use AI applications at Diamond Shadow Productions for:

(a) deep directed research on various things including specific market analysis, financial analysis of studio returns and strategies, technical matters. Some AI systems - to the extent they have the capability to analyze statistics and access to actual data - is getting good at this, but only if you are good at it yourself. It can save weeks of research and can locate and summarize reliable studies, published data on audience reactions, assess the truth of claims on certain industry websites, and much more. (Ironically, I ran a deep analysis of a venture capital funding site that made specific claims about its AI tools - The AI was able to determine the specific kind of AI algorithm and methods needed to do the job, and that those tools could not be created without breaching a host of laws and having access to a ton of private organizations at a cost running into the hundreds of millions. Not an uncommon thing in the AI marketing bubble right now) But you have to understand the difference between Large Language Models, Small Language Models and other methods of machine learning that the AI might have been trained on - because this impacts the inherent accuracy or inaccuracy of results.

(b) enhanced editing tools - as I have mentioned before, chromakey is going the way of the dodo. AI can track and mask out objects, including humans, without green screen and on unbelievable difficult backgrounds. This has reduced certain post tasks from weeks to days or even hours. Color matching, audio level matching, audio cleanup. These tools have been developing for a very long time and are part of the standard editing platforms already (Avid, DaVinci, Premiere/After Effects). Automated transcription and subtitling is a boon and we use it in production constantly.

Of the tools you list, here's my take having worked with and around them and similar AI applications:

1. Script analysis - adds nothing to the value of the process except perhaps saving the time of an outside reader. Outside readers are a large volume studio thing, not necessary at the end of the day even for them except as a filter. The concept that audience reactions to a completed feature can be predicted at all, especially based on a script which has to go through years of evolution and change to get to market, is dubious at best. Script analysis processes are trained on LLMs or SLMs with no real way to access or process hard data, especially speculative data.Therefore, the results might be interesting but you rely on it at more risk than a human assessment. Slated.com has been selling AI script analysis for near a decade now with, to my observation, no results useful to an independent producer.

2. Virtual location scouting - a natural extension of the process. Cannot replace the in-person scout by any means and for a lot or production reasons, but certainly helps rule out locations that are not worth visiting. So saves much time and expense.

3. Automated editing. These tools are either integrated in the standard editorial toolsets already, or are in process of being integrated. They cannot create an actual edit, never will be able to. However they do automated scene detection, and deletion of visual and audio dead spots, etc, so that the logging process is reduced or gone. They can also create an assembly of shots having removed all that dead space - this may or may not be useful depending on the footage. Basically, the role of a junior assistant editor and the transcription department can be reduced or wiped out.

4. Enhanced CGI - refer to (b) above. These are very valuable, but what is really happening with these tools is that the goalposts of what is commercially acceptable and marketable are being moved down the field. What was good a couple years ago, is now considered almost amateur. So while these tools absolutely save time, and properly used can create better looking cgi - expectations are now higher and you must exceed those expectations in the professional world. An example is a recent discussion here on the use of point cloud generation to quicken the process of creating CGI environments and integrating live action or other 3D models within that environment. This process has been developing with advances in AI for a decade at least. However, now that the process is reaching down to a broader range of productions, the expectations for final results are getting higher. This will continue, so that the results will never be achievable by the amateur filmmaker or consumer. Instead of helping those sectors compete, AI actually ensures they will never be able to.

5. Personalized Marketing. This I must confess I am not sure how useful it is or where it will land, legally on the one hand, and for effect, on the other. There are so many variables in marketing that have nothing to do with the filmmaker, and everything to do with the integrity of datasets, adapting to a evolving AI-based search engines, and access to traffic (currently de facto controlled by Google). That it is a thing currently is without doubt.

I am very conflicted....98% of me wishes AI a quick painful demise but 2% understands that history is created with new tools and progress. I am enough of a student of history to recognize previous generations felt similarly about the auto, the telephone, the calculator, the computer, the internet, the cell phone etc.....I am also thoughtful enough to believe that AI is truly different from these past creations and the ramifications for it are much more rapidly occurring in an environment that is increasingly designed to be manipulative, divisive and unregulated. I am pragmatic enough to know it matters little what any of our opposition will really mean - NOTHING. I am pessimistic enough to believe that this is the ultimate transition away from carbon based organic humanity to a synthesis resulting in future generations being a biological computer or a technical human. I hate it and try to avoid it every way possible.

Thanks @ Shadow for your detailed input, the puzzle yields more, regards pros/cons of AI in film production. @Darrell Biological & technological fusion is no longer an Orwellian fantasy, though I’m conflicted too Re AI. Admittedly, it’s superior to controversial prior inventions. LOL They reckoned mobiles fried brains & microwaves had deadly X Rays but will AI pull the total “switch” & clone human machines? No, nothing will.

will i embrace AI as part of the future of filmmaking? ... well it depends, will some someone have a (figurative) gun to my head forcing me to use it?

Robert Bruinewoud and @Darrell Pennington - AI's been a part of filmmaking for a long time. If you use any standard editing system, any current CGI, match-moving, 3d work, color matching, audio processing, transcription software, you're using AI. If you are thinking of the mediocre generative AI of ChatGPT et al, it's a statistical model that predicts words, and except for expansion of vocabulary, it isn't much more impressive than the original Eliza, which was developed in 1962. Those services get all the press because they are consumer-grade products and are marketed to consumers. They will never be good, because they are based on statistical algorithms, which by definition can only produce average results. In other words, they don't think or assess your questions, they correlate words associated with your questions... They don't even seek to emulate human thought which, still at this point, is not understood and which sits within a brain which has more computing power than the entire planet of computing devices (for real) put together. The dangers of AI are only that a sector of the public will begin to think of it as giving them real answers. Currently though, there is already a backlash about the poor quality of generative AI and the inability of AI to give actual reliable answers - so perhaps that isn't a real fear.

Debbie Croysdale Absolutely, Debbie — I see AI less as a “clone switch” and more as a tool to free creatives from the heavy lifting so they can focus on storytelling.

Robert Bruinewoud Haha, fair point Robert — no one should feel forced. Ideally AI should be a choice, just another tool in the filmmaker’s kit.

Shadow Dragu-Mihai, Esq., Ipg

Really insightful. I like how you put it into perspective. Most filmmakers are already using AI without realizing it through editing, CGI, and sound tools. The generative side may have limitations, but it still sparks debate around how much space we should allow it to take in the creative process.

Darrell Pennington , your concerns are valid. In filmmaking, AI can streamline tasks like editing, previsualization, or script analysis, letting creators focus on imagination and vision. Using AI thoughtfully might allow us to guide its influence rather than be controlled by it.

It is certainly possible - I believe that society will dictate the psychology of how AI will be used with what they usually use, the dollars available to them. Much like the music industry was decimated and in many ways eliminated, I believe that AI trumps the power of streaming by a factor of 10 at least. So from the day Metallica sued Napster to today, the psychology was established and now non physical music is now approx. 69% of revenues vs 19% for physical sales. AI will likely dominate industries at a larger % and in a much quicker time frame is my biggest concern.

I believe AI should never replace the creative process. It is not here to take away the artistry that makes filmmaking so powerful. Instead, we see it as an assistant: a tool that handles the repetitive, technical, and time-consuming tasks so that filmmakers can spend more of their energy on imagination, storytelling, and vision.

Whether it is doing research while you are scriptwriting, helping you do rapid storyboarding, automating the breakdown and tagging of your script, or even helping you quickly produce your pitch document or budget structure, AI should be there to support you, not define you.

At Pzaz, our philosophy is simple: AI is at its best when it supports the human voice, not when it tries to imitate or replace it. Just like a great assistant in production, it helps streamline workflows, reduce friction, and create more space for creativity to thrive.

Darrell Pennington, I agree — even I, an ordinary user without experience, was able to create original tracks using the free version of Suno v3.5! You can listen to some sample tracks in my video/audio section.

AI is definitely an exciting tool for filmmaking, and I think its potential is huge — especially for tasks like script analysis, virtual scouting, and post-production efficiency. That said, I see it as a collaborator, not a replacement. The human element — intuition, emotion, and creative risk-taking — is still what makes a story resonate.

I’m curious to explore how AI can help streamline processes while keeping the heart of storytelling intact.

What you all said really resonates — especially the point about AI being a collaborator, not a replacement.

I have seen firsthand how many filmmakers are caught between fragmented tools, endless admin work, and the growing pressure to “use AI” even when it doesn’t always feel intuitive or creative. That’s why we’ve been exploring ways to integrate AI in a way that actually serves the filmmaker — automating the technical grind while keeping the creative flow entirely human.

The goal isn’t to hand control to algorithms, but to give creators more time and mental space to focus on storytelling, collaboration, and vision. If AI can take care of the structure, we can spend more time shaping the soul of the story. Let me know if anyone wants to helping us design the future integration of AI into film making.