There was a post over on Reddit VFX covering the usual “will AI replace some or all of us?” ground, and I think it’s interesting to talk about and would like to, just you know, not on Reddit.
Now, I think AI is 2023 blockchain; it’s a thing that tech is trotting out to hide their utter lack of imagination in the last decade.1. It is not an idea in itself, it’s just a handwave that says, “maybe the computer has some ideas, because we sure as shit don’t”
But for fun let’s say it all comes to pass. AI can make a full movie that is indistinguishable from other movies. Will society consume a movie that has no stars, no directors, no editors, no composers and no army of VFX people in the same manner they consume a Marvel film, or will the inhumanity of it be an obstacle. Will John Cena going on talk shows to promote AI John Cena’s new space adventure ring hollow?
Does the lack of a human being conveying an emotion matter? (and also what are your top 3 movies and top 3 tv shows?2)
1. Tiktok is making an Instagram, Facebook is turning Instagram into Tiktok, All new cell phones are shiny rectangles that have slightly nicer cameras than last year, and all your apps are subscription based. 2. I’m collecting data here, like your phone. Furious 7, 7 Samurai, Akira / Severance, Gurren Lagaan, Dark
I think it’s very possible. It’s already happening on a certain level with social media…virtual influencers with millions of followers.
2.8 million followers… 7.1m views of her music video. Things like this set the stage for the acceptance of it infiltrating other forms of media. It gets interesting when you think about content being created that caters to you. Want a new album by your favorite artist that broke up years ago? Feed it into the system and you can have it. How Artificial Intelligence Helped Make an Experimental Pop Album | Smart News| Smithsonian Magazine.
Maybe that’s the future of entertainment? Want another Star Wars film? Give AI your plot and it’ll make one for you.
Ai will replace everyone, who’s not adapting to the future. Like computers did to oldschool typographs, the internet to mail courier services and so on.
There is no reason why it won’t do so, so we shouldn’t ask whether ai will replace us or not, but more how we can upgrade ourself to go further with it.
(I bet with ChatGPT i would have been able to write these sentences within a much more fluent way)
speaking of mail couriers, I got a letter from Chase bank today. I’d closed my account years ago but there was an imbalance and they cut me a check for the difference.
that difference–carried by a human to my door–was $0.08.
I mean we kid, but this is what I’m getting at above I suppose. How do you future proof (career-wise) against this? If we get to the point where so many traditional jobs across all sorts of industries can be done by a very, very small group of people, who is the one pushing the button? Because I bet it’s not me. And I bet it’s not the guy who spent all his time typing prompts into generative AI. I bet it’s the CEOs nephew…
There is no proof, especially as it’s in the future.
I personally hope advertisements will stay in the same pattern. When everyone prompts the same ideas, based on same models, everything won’t be that different at all (it’s not different for 20 years but let’s skip that point:D)
So now with everything being the same, you need someone to change it to something special and unique.
I honestly think that creative directors and clients need a human on the other end, either to blame, or to feel understood by. If it’s just them, they’re taking a lot of heat. At some point they’ll want their blast shields back and we’ll be promoted to button pushers.
So far, Stable Diffusion and ChatGPT seem really good at producing average-ish things, based on others that humans already came up with.
Humans will always have the advantage of intuition and imagination. AI may be really good at “filler” things like stock images or maybe background music scoring for TV shows -within some very predictable genre- but, imho, won’t be able to break out of these already preconceived boundaries that people have already created. The whole machine-learning process would seem to set that limit.
It took him almost 20 hrs to do this still going through iterations. Photoshop and all. The guy that won the Sony Photo of the year went through a similar process. Not easy getting an exact result.