Is AI the end of photography?

ancient_mariner said:
I suspect a key reason ordinary people don't care about AI creating unreal images is the expectation that photographers have been doing it for years in Photoshop. WE have trained them to accept it.
Click to expand...
I'm not sure because what I also don't understand is why people are so accepting of these LLM AI's being wrong. I think part of the problem is people think these are just issues with new technology and they'll get better but also I think a big part of the problem is many people think these systems are much smarter than they actually are. So they don't realise that the generated images and texts are not the product of some clever and powerful intelligence but something that just more simply been stolen from somewhere else.
 
myotis said:
It's built into the latest Leica cameras, Nikon are introducing it to Nikon cameras and Capture One works with it, Presumably other cameras and software also work with it , but these are the ones I know about.
Click to expand...
I'm perhaps too cynical but it seems to me that there is a simple way around this "authetication"...
  1. Take picture of event with any camera (including "authenticating" camera)
  2. Edit image as desired.
  3. Make suitably sized print.
  4. Photograph print using "authenticating" camera, with date and time set as required.
  5. Look: this was taken as it happened!
 
I agree, even if they GPS tag it, the signal can be faked to claim any location.
 
AndrewFlannigan said:
I'm perhaps too cynical but it seems to me that there is a simple way around this "authetication"...
  1. Take picture of event with any camera (including "authenticating" camera)
  2. Edit image as desired.
  3. Make suitably sized print.
  4. Photograph print using "authenticating" camera, with date and time set as required.
  5. Look: this was taken as it happened!
Click to expand...
Like anything, I am confident there will be ways around it, but I am assuming a big part of the protection will come from controlling who can sign up to the contents credentials cloud and how they police it.

It comes back to what said in my first post, about "trusted sources" and the contents credentials approach seems to offer a way of managing trusted sources and reducing the risks of fake content.
 
jerry12953 said:
Where as what we will get is actually the opposite. AI will do the fun stuff and leave us to do the chores!
Click to expand...
As per my previous post. AI is really good at doing the complicated boring stuff (all the complex maths stuff) but rubbish at the things a human finds easy.
So that’s why they’re producing amazing models of complex objects which can be ‘lit’ from any direction. But completely baffled by capturing genuine human emotion, sports, etc.
 
Phil V said:
As per my previous post. AI is really good at doing the complicated boring stuff (all the complex maths stuff) but rubbish at the things a human finds easy.
So that’s why they’re producing amazing models of complex objects which can be ‘lit’ from any direction. But completely baffled by capturing genuine human emotion, sports, etc.
Click to expand...
I agree. I'm finding Gen AI great at things I find boring
 
wilt said:
I recently seen Coca Cola created a AI video "Holidays are coming" ad and it's terrible. This is unfortunately where the industry is headed, the cheap and do rightly attitude.
Click to expand...

That was rightly slated in the ad industry too.

There's a Vodafone ad that's also entirely AI which is much better, but for reasons that have been mentioned in this thread. It's a series of very loosely connected vignettes, so you don't have to have the continuity throughout that you might otherwise have in an ad. Non AI generated ads also use this too - and often for the same reason. They're using stock library footage and piecing it together.

View: https://youtu.be/9AyEC_K9kBg?si=OrE41cx49hO1QLqx


Someone mentioned earlier about food photography. Again, speaking from experience what tends to happen is that big companies (McDonald's, Burger King, Pizza Hut etc) have to show actual product. Cooked in the correct way. The food stylists will come in and arrange it to make it look appealing, but they're under pretty strict regulatory bodies to ensure it's 'genuine'. i.e. you can't use a half-pound patty if the actual burger is only quarter-pound. So those videos that used to do the rounds on Insta and the like of food photographers using PVA glue instead of mozzarella, or soy sauce for coffee, would only apply to generic photo shoots for stock libraries rather than for a brand.

And it would be independent places - especially fast-food - that would use this type of photo. They'll just have generic pizza/burger/wings shots which is where I could see AI images being used. Proper restaurants, even independents, will always want to showcase their food, so I'd be hopeful anything you see will be commissioned photography.

I sort of feel like it's coming whether you like it or not (I don't).

But it's not a new thing. Pretty sure people weren't that impressed with the Spinning Jenny either. I also imagine that a load of painters may have been sitting around 200 years ago worrying about whether photography was going kill painting. It hasn't entirely, but it's not the industry it was.
 
jerry12953 said:
Does it do the washing up though?
Click to expand...
No, would be great if it did though :)

I live-streamed a conference a while back that included a session on physical AI bots. The example case discussed was their use to provide companionship to terminally ill children in hospitals in France. Bots have more time for the children than human nurses and the children form strong relationships with the bots. It made me feel nauseous, actually, that bean counters in the hospitals prefer this option to more human care.
 
You must log in or register to reply here.
Back
Top