AI is here, but will it really replace us when it comes to video editing?
We’re not even halfway through the year yet, but 2023 is already the year of artificial intelligence (or AI as it’s more commonly known today).from AI-Powered Products and NAB Updates To New short films and other AI production projects Once on the Internet, there are many things to follow and monitor. Especially if he’s worried that eventually he will lose his job to AI.
While I personally would argue that AI is likely to be used more to speed up workflows and reduce tedium in most productions, there are many different film and video professionals who believe that AI is There is an imminent threat of being completely replaced.
one of the first arenas AI is about to move to video editing. But how real is this threat today? More importantly, how AI may not be able to fully replace human editors (and the few studies that prove it) Is there
1. Real human emotions
according to Article published by MIT, AI is trying to tackle human emotions when we speak. The magazine says:[W]Humans may currently have the upper hand when it comes to reading emotions, but machines are using their strengths to gain ground. ”
But unlike other forms of AI that can gather a lot of information quickly, AI simply lacks a lot of data and struggles to keep up with emotions. Most notable is the machine learning aspect of human voice and facial cues. These are some of the only indicators of human emotion that machines can track.
others, like bimo putro tristianto, “Despite all these advances, it is important to note that AI systems are still far from being able to feel emotions in the same way humans do. Maybe, but they don’t have the complex psychological and physiological responses of humans.”
This is because the idea that AI can reproduce authentic human emotions, and use that information to make complex decisions in the video editing process to communicate these emotions, has been around for quite some time now, and probably in the future. means to continue. It seems very likely.
2. Dramatic and comedic beats
Using some of the same logic as above, without the same sense of real human emotion at its core, the AI would be able to recreate these human emotions in its compilation of both dramatic and comedic moments. You will constantly struggle with the AI of In particular, I personally argue that comedy is perhaps the most difficult thing for AI to truly recreate.
Comedy is difficult, even for humans, and highly subjective to some. But as an audience as a whole, we find it funny when we watch it. Humor stems from a deep understanding of the human experience, subverting expectations, challenging social norms, and requires pinpoint control to strike a perfectly timed punchline and joke.
3. Ironic Moments
Like comedy, irony can be difficult to explain to other humans learning about it for the first time. It may even be difficult to explain to machines and algorithms. Irony, in particular, has human experience embedded in it, and it may be very difficult (if not impossible) to connect specific facial cues and other data AI can track to his points. .
It’s very difficult to create sarcasm in editing, but experienced editors are always making comedic and dramatic decisions, using these elements that AI alone might not be able to quickly create to create stories. helps to convey
4. Free association and improvisation
The same applies to free association and improvisation-based stuff. In a sense, improvisation itself is a unique form of human algorithmic performance. A person receives a prompt, replaces it with their own human experience, and presents the same information as before, but that information changes based on themselves and their creative process.
For video editors, there is more free association and improvisation than you might think. The amount of loose improvisation an editor needs to do is almost limitless, as there is clearly no single path to take when taking footage from a shot and putting it together in a video. AI can make many similar decisions, but its complex algorithms are no more unique (that is, more human) than human algorithms.
5. Natural Timing
Finally, and while this might be a bit debatable, natural timing is probably always going to be a struggle for AI, and timing is a big part of video editing. However, it can be difficult to explain to someone new to editing what a “beat” is and how to create a beat in editing.
There are strict rules about how many frames before and after a cut should be considered, but even the strictest video editor will tell you that these rules are all subject to interpretation and can be changed at any time based on the editor’s sensibility. . The natural timing of the piece.
When you’re messing around with many scenes with complex timings in your sequence, a difference of a second or two at a time can make a big difference in how a scene feels and how your audience perceives it.
But what do we know? As you know, this article was written by AI.
So who can say what will actually happen in the future? Still, let us know your unique human thoughts on how AI works in the world of video editing in the comments below.
