That the internet as we knew it no longer exists is not a surprise: it has been filled with search results generated by artificial intelligence and ‘slop’. The consequences are already visible: clicks have been reduced by half, which is catastrophic for the media. But it’s not just the text that is suffering from this barrage of AI that blurs everything: we no longer know how to distinguish whether an image is real or not, we have gone from documenting our lives on social networks to the era of influencer content favored by the algorithm to videos and images that are not real, but can pass as such. There are no longer four fingers that are worth it.
Instagrammers, the feed is dead. And this is also going to take its toll on social networks. Adam Mosseri, CEO of Instagram, closed 2025 with a publication in the form of a presentation of 20 images where he reflected in depth on what is coming: “the era of infinite synthetic content”, the antithesis of a more personal Instagram that has been dead for years.
For Mosseri, AI has turned the carefully maintained grid with its algorithm into something of the past: “Unless you are under 25 years old and use Instagram, you probably think of the app as a feed of square photos. The aesthetics are careful: a lot of makeup, skin softening, high-contrast photography, beautiful landscapes,” Mosseri’s sentence falls like a stone on this millennial, who still uses Instagram as a kind of photo album. “That feed is dead. People largely stopped sharing personal moments on the feed years ago.”
Tap to go to the post
In search of something real. Mosseri explains that now its users keep their contacts up to date on their personal lives with “improvised photos of unflattering shoes and poses” shared via DM. And this also affects content creators: the omnipresence of images made by AI is going to bring a change: goodbye to those pro-looking photographs in favor of a more real and improvised aesthetic: “Flattering images are cheap to produce and boring to consume. People want content that feels real.”
In fact, the CEO of Instagram points to manufacturers, applicable to cameras and mobile phones, who he says are making a mistake by democratizing the ability to “look like a professional photographer from 2015.” Because RAW images with defects are still a sign of reality until AI is able to copy them.
But what is real? The time has come to unlearn to believe what our eyes see, something we have been doing all our lives. Javier Lacort explained that our entire epistemology (which ranges from judicial testimony to photo albums) is based on the fact that seeing is a way of knowing. If you see a tiger, there is a tiger. If you see a photo of a tiger, someone has been close to one. This no longer applies: the era of uncovering organized fake news has given way to anyone with Nano Banana Pro being able to get such an absurdly realistic image with a basic prompt in a few seconds. Now creating a deepfake is trivial.
Adam Mosseri thinks the same. “For most of my life I was able to safely assume that photographs or videos were largely faithful captures of moments that actually happened. That’s clearly no longer the case, and it’s going to take years to adjust. We’re going to go from defaulting to assuming that what we see is real to starting from skepticism. To paying attention to who’s sharing something and why. This will be uncomfortable: we’re genetically predisposed to believe our eyes.”
If you can’t beat them… The paradigm shift has already occurred, so now Instagram and other platforms have to adapt to this new reality: “we have to build the best creative tools. Label AI-generated content and verify authentic content. Show credibility signals about who is posting. Continue to improve the ranking of originality.” It is the apocalypse of what is a photo that we have been predicting for years.
Focusing on Instagram, Mosseri talks about how “we like to complain about ‘AI junk content,’ but there is a lot of amazing content created with AI.” He doesn’t give concrete examples or talk about Meta tools to make this possible, but Meta has already added AI tools on Instagram and Facebook. Without going any further, its AI Studio allows you to create personalized chatbots to deal with your followers.
New times, new identification measures. It is increasingly difficult to identify content in AI, so it proposes fingerprints and cryptographic signatures in cameras to identify real content, forgetting about labels or watermarks. In any case, it advocates greater transparency about who publishes on the platform and improve creativity so that its human users can compete with content made in AI.
In WorldOfSoftware | The future of the Internet is to be flooded with AI. And there are those who have already seen a business niche: content made by humans
In WorldOfSoftware | There is a generation working for free as a documentarian of their own life: they are not influencers but they act as if they were.
