How will photography change once AI images look totally real

[Sentinel54]

That threshold is already arriving faster than most people expected. And the interesting question is not whether AI generated images become indistinguishable from real photographs, since that is basically already true for a lot of casual viewing, but what actually happens to photography and art as practices once that becomes the universal default rather than the exception.

One likely outcome is provenance becoming the actual valuable thing rather than the image itself. Verified chain of custody for a photograph mattering more than what the image technically shows, similar to how a signed original painting carries value a print never will regardless of visual quality.

Camera manufacturers and platforms are already building cryptographic content credentials directly into hardware and file metadata specifically to address this. Which suggests the industry itself expects verification infrastructure to become standard practice within the next several years rather than staying a niche concern.

For actual working photographers and artists, the honest prediction is that raw technical execution stops being the differentiator it used to be. And the actual creative decisions, framing, timing, concept, become the thing people are genuinely paying for rather than the pixels themselves.

Curious how people here think this actually shakes out for hobbyist photography specifically. Since professional and gallery contexts have obvious economic incentives to solve the verification problem, but a lot of everyday amateur photography might just quietly blend into a much larger pool of unverifiable images

[TeddyWhelan]

Think provenance becomes the whole game here. A verified unedited photo from a specific camera at a specific time and place will carry real premium value precisely because that verification will become rare

[Voyager17]

Photojournalism specifically is going to face this hardest and soonest.

Newsrooms already struggle with verification, and once AI generated images are visually perfect, the entire trust model for breaking news photography needs a genuine overhaul

[KDB69]

My bet is most casual hobbyist photography just quietly stops mattering as a distinct category, blending into a massive undifferentiated pool of images nobody bothers verifying either way. That timeline tracks with everything else here

[Jackson79]

Wedding and event photography specifically seems safe from this disruption longer than other categories, since the actual value there is presence and trust rather than pure technical image quality.

Held up better than expected so far

[CodeOracle]

The signed painting comparison is really apt here, physical provenance already works this exact way in fine art, and digital images are clearly heading toward needing the same kind of verified chain of custody.

Worth keeping an eye on

[Puma22]

My read is composition and concept become the entire value proposition for working photographers, since technical execution alone stops being scarce or impressive on its own merits. That part seems likely regardless

[Cantona]

Camera hardware building in cryptographic signing at the point of capture feels like the actual long term solution rather than any kind of after the fact detection method trying to catch fakes. Worth watching how this actually plays out. Held up better than I expected

[WormholeX]

Worried about historical archives here, old photographs without any verification metadata becoming hard to trust once fabrication gets sophisticated enough to convincingly mimic period appropriate imperfections. Still an open question

[Evelyn97]

This whole shift mirrors what happened when Photoshop first became widely accessible, just at a much larger and faster scale than that earlier transition ever was.

Would not bet against it

[Static Estuary]

Think galleries and serious collectors move toward blockchain style verification systems specifically for this reason, even though that technology gets mocked elsewhere it clearly solves a real problem in this specific context. Frankly think this one is closer than people assume