Adjusting lighting, blurring backgrounds, intensifying looks: artificial intelligence can do everything photographers do when it comes to retouching... without the user even realizing it, thanks to features on smartphones.
Last month, an amateur photographer with the handle "Ibreakphotos" decided to run an experiment on his Samsung smartphone to find out how a feature called "spatial zoom", launched in 2020, which claims a 100x zoom rate, really works.
Samsung, the South Korean manufacturer used sparkling clear images of the moon to promote the feature. "Ibreakphotos" took its own shots of the star, blurred and without detail, letting its phone add craters in particular.
The built-in artificial intelligence software used its "training" data on many other photos of the Moon to add detail where there was none.
"Samsung's photos of the Moon are fake," the amateur photographer concluded on the Reddit forum, leading many to question whether the shots they take are really theirs and whether they can even be considered photographs.
"I wouldn't say I'm happy with the use of AI in cameras, but I'm okay with it as long as it's clearly communicated what each processing channel actually does," "Ibreakphotos" told AFP, asking not to use its real name.
Samsung has defended its technology, saying it does not "overlay" images and that users can disable the feature.
The company is not alone in the race to integrate AI into its smartphone cameras: Google's Pixel and Apple's iPhone have been marketing such features since 2016.
But the AI debate is not limited to amateurs on discussion forums: professional organizations are sounding the alarm.
According to Michael Pritchard of the British Royal Photographic Society, the industry is being inundated with AI, whether it's in cameras or software like Photoshop. "This automation is increasingly blurring the boundaries between a photograph and a work of art," he tells AFP.
The nature of AI is different from previous innovations because the technology can learn and bring new elements beyond those recorded by the film or sensor. This opens up opportunities but also "fundamental challenges to redefine what photography is and how +real+ a photograph is," he adds.
Of greater concern to professional photographers, however, is the rise of AI tools that generate entirely new images from text, such as DALL-E 2, Midjourney and Stable Diffusion.
"This is not authorship" and, "in many cases, it is based on the use of unlicensed work training datasets," notes Nick Dunmur, a member of the British Photographers' Association, while the phenomenon has already led to lawsuits in the United States and Europe.
"AI will not lead to the death of photography," he believes. A prediction shared by Michael Pritchard, who reminds us that photography has endured from the daguerreotype to the digital age and that photographers have always risen to technical challenges.