Posting an AI service on IndieTalk...
- Thread starter indietalk
- Start date
I guess the logic escapes me, We're using ai to make filmmaking cheaper and more accessible. You're angry because you think it'll give giant corporations the edge. Giant corporations have the edge right now, and the main reason is that it takes $100 million to make a movie.
Crashing that price and making filmmaking accessible to everyone has nothing to do with the quality. Sure a lot of people will make junk, but how is that any different from the way it's always been? Junk writers will make junk scripts into junk movies, you think once the cinematography is all the same quality we won't be able to tell the difference?
Think about what's actually changing here, The price of entry. Now an indie filmmaker will actually be able to make a film. Before the ai revolution and products like I'm developing, only the super rich were ever even allowed to try to make a film of a quality audiences might accept. Now anyone will at least be able to attempt it instead of being gatekept by this $100 million price tag. Whether the quality of film X is good or not has very little to do with the tools used to make it, beyond audiovisual standards which will soon be equalized by the emergent technology.
The real quality of a film lies in its story it's emotion and characters, Its plot and pacing, Its unique creative spark, all things that ai isn't responsible for, and doesn't do. So we're not really talking about a change in the quality of movies as much as we're talking about leveling the playing field so that any person who has a worthwhile story to tell can tell it without having to sit and wait for their wealthy overlords to give them permission to do so. Maybe you think quality comes from the refinement process available through only allowing writers with $100 million worth of Hollywood friends to publish. I think we get a bunch of nepotism films and pop star autobiographies from that. I think we get bland ideologies disguised as motion pictures and nieces and nephews disguised as movie stars. What if we just let the people decide instead?
Why should independent creators be barricaded behind a financial wall? Why not move towards a world of free and open competition wherein the best creatives win regardless of their financial situation?
Crashing that price and making filmmaking accessible to everyone has nothing to do with the quality. Sure a lot of people will make junk, but how is that any different from the way it's always been? Junk writers will make junk scripts into junk movies, you think once the cinematography is all the same quality we won't be able to tell the difference?
Think about what's actually changing here, The price of entry. Now an indie filmmaker will actually be able to make a film. Before the ai revolution and products like I'm developing, only the super rich were ever even allowed to try to make a film of a quality audiences might accept. Now anyone will at least be able to attempt it instead of being gatekept by this $100 million price tag. Whether the quality of film X is good or not has very little to do with the tools used to make it, beyond audiovisual standards which will soon be equalized by the emergent technology.
The real quality of a film lies in its story it's emotion and characters, Its plot and pacing, Its unique creative spark, all things that ai isn't responsible for, and doesn't do. So we're not really talking about a change in the quality of movies as much as we're talking about leveling the playing field so that any person who has a worthwhile story to tell can tell it without having to sit and wait for their wealthy overlords to give them permission to do so. Maybe you think quality comes from the refinement process available through only allowing writers with $100 million worth of Hollywood friends to publish. I think we get a bunch of nepotism films and pop star autobiographies from that. I think we get bland ideologies disguised as motion pictures and nieces and nephews disguised as movie stars. What if we just let the people decide instead?
Why should independent creators be barricaded behind a financial wall? Why not move towards a world of free and open competition wherein the best creatives win regardless of their financial situation?
When I worked in foreign exchange, we had a phrase: "talking your book."
It meant that a trader expressed their opinions on the currency market in a way that they hoped would motivate others to take positions that
would increase the value of their holdings.
For example, if they were long euros vs the dollar, they'd talk up how undervalued the euro was, where the technicals show a bounce, etc.
I respect the work you're trying to do @Nate North but you'll never convince me that AI is anything other than another way to make cheap shit cheaper and faster and make it easier for people in power to steal from others who actually do good, original work.
It meant that a trader expressed their opinions on the currency market in a way that they hoped would motivate others to take positions that
would increase the value of their holdings.
For example, if they were long euros vs the dollar, they'd talk up how undervalued the euro was, where the technicals show a bounce, etc.
I respect the work you're trying to do @Nate North but you'll never convince me that AI is anything other than another way to make cheap shit cheaper and faster and make it easier for people in power to steal from others who actually do good, original work.
Last edited:
Not for the first time, Nate, I disagree with this tendency of yours to drag everything back to a question of dollars. Sure, there is a financial aspect to the "price of entry" - as for any aspect of the creative arts - but the real "price" of being able to make a movie is being able to create a team - a group of artists and technicians that can work together to turn an idea into a viewable product, and then a network of sufficiently interested parties that can distribute that work to a receptive audience. Yes, every step of that process can be automated and/or technologified to an extent, but the more you remove "miscellaneous others" from the endeavour, the harder it is to produce something successful and have it seen by those who would most appreciate it.
The evidence is available all over YouTube, where "content creation" has become little more than one person yakking into a microphone about their favourite topic, accompanied by a reel of random stock video clips with the most tenuous link to the subject matter. Even when a "content creator" puts some real effort into the creative side of the equation, lowering the "price of entry" means we're treated to presentations by people who can't turn off the auto-focus on their camera, or who can't seem to deliver a whole sentence without fluffing their lines ; and worse still, you'll find they've two, three or four years' worth of the same bad practice under their belts and have never bothered to improve.
No. The real quality of a story lies in its emotion and characters. That story can be told through the written word and read from a book, or by a disembodied voice heard through headphones, or be acted out and displayed on a stage or a screen. A film is way, way more than the story alone. If every aspect of lighting, camera angles, sound design, etc, is handled by an "AI" assistant that does exactly what it's told to do, then you've reduced the whole production process to one all-powerful, don't-question-my-artistic-vision director's cut. That's not a recipe for success. All it does is flood our viewing platforms with hours and hours and hours of crap that crowds out the good stuff.
THIS exactly.
A few thoughts on AI in general:
1 It's here. Get used to it.
2 Take a moment to appreciate that it is something new and extraordinary on the face of the earth. A mundane example: The greatest chess player that has ever been is not a person, but some variety of toaster, that taught itself to play in one afternoon, starting only with how each piece moved.
3 It can help. It is not impossible, probably not unlikely, possibly inevitable, that whatever combination of luck, insight, technology, and dedication combines to make medical breakthroughs--maybe Parkinson's, maybe spinal choard injuries--can be replicated by AI. Assume that a way to repair spinal choard injuries potentially exists. Assume that finding it is a one in ten billion chance. For any group of hu-mahns those are long odds, but for the AIs it's no sweat. A days work? An hours? It's easy to imagine some StarTrekky future when Parkensons's Disease is a thing of the past, where spinal cord injuries are routinely repaired. Before defecting on AI itself consider: this future could be 2026, not 2626. If some AI helped to discover a way to reverse Parkenson's, would we really worry about it putting doctors out of work?
4: Also, of course, it's not impossible, probably not unlikely, possibly inevitable, that some AI will decide, maybe on some global all-of-humanity scale, that it didn't wan't to open the pod bay door after all. Yea there's that. Oh well.
1 It's here. Get used to it.
2 Take a moment to appreciate that it is something new and extraordinary on the face of the earth. A mundane example: The greatest chess player that has ever been is not a person, but some variety of toaster, that taught itself to play in one afternoon, starting only with how each piece moved.
3 It can help. It is not impossible, probably not unlikely, possibly inevitable, that whatever combination of luck, insight, technology, and dedication combines to make medical breakthroughs--maybe Parkinson's, maybe spinal choard injuries--can be replicated by AI. Assume that a way to repair spinal choard injuries potentially exists. Assume that finding it is a one in ten billion chance. For any group of hu-mahns those are long odds, but for the AIs it's no sweat. A days work? An hours? It's easy to imagine some StarTrekky future when Parkensons's Disease is a thing of the past, where spinal cord injuries are routinely repaired. Before defecting on AI itself consider: this future could be 2026, not 2626. If some AI helped to discover a way to reverse Parkenson's, would we really worry about it putting doctors out of work?
4: Also, of course, it's not impossible, probably not unlikely, possibly inevitable, that some AI will decide, maybe on some global all-of-humanity scale, that it didn't wan't to open the pod bay door after all. Yea there's that. Oh well.
Well said, Spike.
I asked an AI (chat GPT) what it thought about my defense. It wasn't that impressed, lol.
I forgot to add: the thing is kind of fun. I can admire how, along side its kind of amusing (to me) sententiousness, it can get things like tone and humor and respond, like the "open the pod bay door" reference, or me calling it a toaster.
I forgot to add: the thing is kind of fun. I can admire how, along side its kind of amusing (to me) sententiousness, it can get things like tone and humor and respond, like the "open the pod bay door" reference, or me calling it a toaster.
Last edited:
What we want: AI on phone that can change any settings with verbal command, make phone calls for me and stay on hold for hours waiting so i don't have to, screen my phone calls for spam before I answer them
What we get: AI that sticks its nose as far up your ass as possible to sniff out every single byte of data on your personal life
What we get: AI that sticks its nose as far up your ass as possible to sniff out every single byte of data on your personal life
I've tinkered with AI mastering (of songs) and have been impressed with some of it. I'm sure a lot of it is presets and algorithms but they do coin it AI mastering. If you want a quick master of a song fast, it's great. For example you may home record a song and mix it yourself and want it to have that mastered sound. You can upload it and in a minute or two it spits out the song for you to preview, and there's a few options each time to run it through to get what you want. If you don't like it, sure, do it the real world way. I've always had my stuff professionally mastered.
Mastering is an art, and it usually takes a visit to the studio (with acoustics designed speficically for mastering) to sit in on the session (for those that care) and a chunk of change. But AI lets some of us quickly master something, something that may not even be ready for the studio. A demo or practice. Example: Perhaps my band played a show and I recorded it with my phone. Now I can take that video file, upload it, and it will spit out a better version of the sound (clearer, better bass, better treble, better volume, etc.) for free (some sites) or subscription. Honestly, it has its place. And I'm not going to pay a mastering engineer to master a file from a concert or rehearsal studio. That's not what they are for. In fact they'd laugh or tell me it's not worth mastering.
As far as using it for anything pro, I have not yet but I have tested it and some of it is quite good. There's already artists using it. Especially those that home record. You may have even heard a song and not known it.
See, now I'm a target for the audio guys on here.
Just a case study type post for reference.
Mastering is an art, and it usually takes a visit to the studio (with acoustics designed speficically for mastering) to sit in on the session (for those that care) and a chunk of change. But AI lets some of us quickly master something, something that may not even be ready for the studio. A demo or practice. Example: Perhaps my band played a show and I recorded it with my phone. Now I can take that video file, upload it, and it will spit out a better version of the sound (clearer, better bass, better treble, better volume, etc.) for free (some sites) or subscription. Honestly, it has its place. And I'm not going to pay a mastering engineer to master a file from a concert or rehearsal studio. That's not what they are for. In fact they'd laugh or tell me it's not worth mastering.
As far as using it for anything pro, I have not yet but I have tested it and some of it is quite good. There's already artists using it. Especially those that home record. You may have even heard a song and not known it.
See, now I'm a target for the audio guys on here.
Just a case study type post for reference.
It's funny how Google doesn't allow you to turn off the AI Overview feature at the top of results. And man this thing is wrong so often. I've googled things like advice on taking care of certain plants and I know the AI is wrong so I will look where they got it from, and on that page it will be advice for a totally different plant. BUT something about the plant I searched is on that page too. So it just scans the page and makes a mashup of it all. IMAGINE that was a medical search, and people stopped at the AI Overview! No really. This is happening. This is not the "good" AI.
And you wonder why I hate AI?
BBC.com has been reporting on (and I've posted about) AI can't even scan and summarize an article correctly.
It's doing the work of humans cheaper and faster, not better.
Last edited:
I never wondered
I keep getting ads in my feeds for some crappy plug-in for NLEs that claims to take care of the audio mix for your video or film, completely AI-driven. I’m not buying it, and I don’t mean that in the purchasing sense but in the accepting that it’s a viable tool sense.
Yes, it’s all going to be based on algorithms. It only knows where to place any one track in the mix based on which other tracks are present. Thing is, there’s absolutely no human touch to any of that. There is so much emotion that can be crafted into a re-recording mix that it cannot be distilled down to a formula. I dare say that even things like corporate videos need to have at least a bit of that emotive touch to the mix.
But I fear that more editors will lower the bar, accept the results of that plug-in, and call it good enough. Hey, it saves them time and money, right? And that’t the most important part, right? Art is all about cost-cutting, right?
I hear Leonardo da Vinci actually set up a team of automatons to paint the Sistine Chapel.
Oh hell no I would not trust it to mix anything. The mastering AIs though are really algorithms like you say and just enhancing things, like an "Auto Fix" button in a photo editor instead of doing it yourself. I mean if I can take my crappy rehearsal recording and before I send it to my band members, run it though this and it sounds better, I'm going to do it. But I'm not sure how much AI that is much as it is presets and algos. I think they are cashing in on the term. These enhancers have been around forever. Now they carry a stigma that isn't necessary.
If you want to try a free one try https://www.bandlab.com/mastering
But one I liked better was https://www.waves.com/online-mastering (still free to try)
But one I liked better was https://www.waves.com/online-mastering (still free to try)