The old Politburo would love this. So will the new dictators.
A few text commands and trump is fit and trim. Epstein has been disappeared. Jesus smiles on everyone.
Google has updated its Gemini AI image generation tool with a build that caused a stir after it was released under the code name Nano Bananas. The upgrade, technically called Gemini 2.5 Flash Image, lets users generate images through voice and text prompts, including swapping out participants in a photo, changing what they are …
In order to make that happen, Trump would have to take a page from the Politburo playbook and no longer appear in public so we're only shown pictures/video of him. His ego wouldn't accept being locked away in the White House spending the next 12 months (if you believe Steve Bannon) rotting away like a Dorian Gray painting not getting the public adulation he so desperately needs.
Years ago Matt Stone and Trey Parker made "Sassy Justice", which was all AI generated. Essentially had Donald Trump as a news reporter and it was talking about deepfake AI videos etc.
They had an AI generated Michael Cain voiced by Peter Serafinowicz. Genuinely you could not tell whether it was Cain or not. It did weird me out, still does to be honest. So every time I see anything online now there is an element of that video in my head where I wonder if it's true or not.
Actually, it is about how still and moving images are just about to loose their value as proof or evidence of reality.
This is something that I find bloody scary. How will we in the near future be able to say what is real and what is made up (your AI slop).
The jump from there to the worlds most destructive and reality denying "politician" is a short one, The man and his base IS using AI in their attempts to spin a reality that suits them.
He is also making sure that EVERYTHING is about him. This does not require the so called "TDS".
I was just hoping that something good could come out of this deniability, but I do not have great hope.
Thankfully there is no possibility this could ever be used for any sort of harm...
As someone who did image segmentation for my dissertation some decades ago what can be done now is bloody impressive. Sadly I'll never dabble in it because I never feel the need to tinker with my photos. They are a record of the reality that prevailed at the time, warts and all. Perhaps the general population could get on board with that concept -- real life, that is.
I'll extend that to dust spot removal, and some judicious cropping - particularly after this weekend when from somewhere in my camera in the middle of a session a bit of dust mysteriously landed on my sensor, and rather than the traditional round blob had managed to form itself into the shape of a cock and balls. I suspect next weekend will involve a certain amount of sensor cleaning.
The only tinkering I've done generally is on group photos at events where typically one or two people missed being at the shoot - the organisers invariably want you to edit them in. It's bending reality slightly but I wouldn't edit in someone who wasn't at the event at all.
Otherwise yes, brightness, contrast, colour, etc. do get tweaked to compensate for the lighting conditions.
As a semi-pro photographer all I can say is, this is the same of sort of shit that people who say "Photography is just pointing a camera and clicking a button! I can shoot better pictures on my phone!". Yeah, like saying you can code better than a 30+ year C veteran after watching a few YouTube beginner vids about Python!
Despite AI I'm still earning money from selling my images I shoot for fun doing my favourite hobby!