Open Instagram, Facebook, or TikTok this week, and you will quickly notice that half of the internet has decided to travel back in time to the 1980s.
Your daily feed is likely overflowing with soft, glamorous studio lighting, thick retro sweaters, and grainy film textures. Friends, family members, and old colleagues have smoothly transported themselves into a decade that many of them barely remember, and some never even lived through. Yet, despite the vintage clothing and the old-school hairstyles, they all look exactly like themselves.
Let us be entirely clear right up front. There is absolutely nothing wrong with enjoying this. Artificial intelligence (AI) is an incredible, highly creative tool. Using it to generate a nostalgic, vintage portrait is a genuinely fun way to engage with new technology. The problem that we face is not the AI itself. The real problem is the careless, almost automatic way that we trade our permanent personal data whenever a new social media trend goes viral.
The method behind this digital time travel is quite simple. You upload a clear, well-lit, front-facing photograph of yourself to an AI image generator. You type in a prompt asking the machine to make you look like a 1980s high school portrait. But there is a catch. You must explicitly tell the computer to keep your face recognisable. And, that is exactly where innocent, viral fun collides with a permanent, unavoidable risk.
The public figure exemption and the rest of us
When privacy warnings like this one circulate, there is a common, understandable pushback. Many people argue that they already have hundreds of photos online, so, one more image will not make a difference.
If you are a prominent internet influencer, a highly visible public figure, or someone who routinely posts dozens of high-resolution selfies every single month, this warning might not feel like a pressing concern. And frankly, that logic makes perfect sense. If your face is already plastered across the public internet from every conceivable angle, your digital footprint is heavily established. For those individuals, the data ship has already sailed.
But, for the average, everyday person, the situation is completely different.
When you join one of these viral loops, you are not just posting a blurry, candid photo taken at a crowded family barbecue. You are taking a deliberate, perfectly lit, front-facing scan of your face and handing it directly to a technology company. You are voluntarily giving them a high-quality map of your unique physical features.
What actually happens to your face once you press that upload button? If you are using a well-known platform like Chat Generative Pre-trained Transformer (ChatGPT), the rules depend heavily on your settings. OpenAI notes that regular, personal accounts operate under specific guidelines. Unless you take the time to dig deep into your privacy settings and actively turn off "model training", your interactions and the photos that you upload can be used to teach their systems.
Furthermore, simply deleting a chat from your screen does not instantly wipe your photo from existence. A digital copy can remain stored on corporate servers for a much longer period, hidden away in a data library that you cannot easily see.
The third-party trap
However, major platforms like ChatGPT are rarely the only applications involved in these moments. Viral trends breed intense opportunism.
While the biggest tech companies face a certain level of public scrutiny and government regulation, thousands of unknown, third-party apps and websites pop up overnight to cash in on these fads. In many cases, these small, specialised applications actually do a much better and faster job of creating those specific vintage filters than a general AI chatbot does.
Because a trend is popular and our friends are doing it, our collective guard completely drops. We do not stop to think twice. We rarely pause to check who actually owns the app that we just downloaded, what country it is based in, or how the developers plan to use our photos.
Many of these obscure apps hide sweeping, aggressive rules deep inside their fine print. By tapping "Agree", users frequently give these companies permanent, legal permission to store, alter, or even sell the high-quality facial data that they just uploaded. Sometimes, these apps are built specifically to collect biometric data to sell to marketing firms or other tech developers.
Enjoying the capabilities of AI is a great thing, but, blindly feeding your face into an unvetted, temporary app just to join a fleeting trend is a highly dangerous game.
A shadow economy
Why does a clear photograph matter so much to these companies and developers? Because there is a massive, highly profitable black market for ordinary faces.
It is important to note that men are absolutely vulnerable in this new digital landscape. Deepfake technology is increasingly being weaponised against men for financial crimes. Scammers use clear photos and video snippets to create highly convincing voice clones and facial replicas. These tools are then used for corporate cyber fraud, tricking businesses out of millions, or for personal extortion scams. Both men and women are broadly vulnerable to having their digital identities hijacked by bad actors trying to make a quick profit.
However, when it comes to the most psychologically damaging and invasive corners of this technology, the burden is not shared equally. The darkest parts of the AI shadow economy disproportionately target women.
According to reports from the European Commission, fabricated pornographic material makes up roughly 98 per cent of all deep fakes currently online, and these overwhelmingly feature women. Applications specifically designed to digitally remove clothing from photos of real people have grown from obscure dark-web novelties into a booming, mainstream industry.
These malicious tools need a very specific type of photo to work effectively. They require a single person, well-lit, staring directly at the camera, with their face clearly visible. It is the exact, precise type of photo that the 1980s trend asks and other trends just like it ask you to upload.
Children face an even darker, more complex reality. Because they are still growing, developing, and learning to navigate the world, they are uniquely vulnerable to online harassment and cyberbullying. The scale of the abuse is staggering, and research indicates that a shocking 17% of synthetic sexual content online now involves a minor.
Where do bad actors get these high-quality photos of kids? They almost never execute sophisticated hacks into secure bank accounts or private computer networks. They do not have to. They simply take ordinary, innocent pictures that are already sitting publicly on social media. Often, these are photos happily posted by the kids themselves, or worse, over-shared by proud, well-meaning parents, a phenomenon known as "sharenting."
The true cost of free nostalgia
Treating all participation in AI photo trends as simple foolishness is not helpful, nor is it fair. Technology is meant to be explored. It is absolutely vital to admit that the risks of this exploration are not equal for everyone.
For a powerful executive or a famous celebrity, adding one more high-resolution photo to the internet likely changes nothing about their daily life or safety. But, for a middle school student, or for a young woman living in a small town where reputation matters deeply, the risk is much higher. The social and psychological consequences of having an image manipulated can be devastating.
The fun 1980s trend does not create these bad actors, and the people making the filters are not necessarily malicious. But, participating in the trend makes a bad actor's job incredibly easy. It hands them a perfect, high-resolution map of a human face, posted publicly on a social media profile, and usually labeled directly with the person's real name.
If a password leaks in a massive corporate data breach, you can simply log on and reset it. If an identity card, driver's license, or credit card is stolen, you can make a phone call, cancel it, and wait for a new one to arrive in the mail.
A face cannot be reset. You only get one face for your entire life. Once its precise measurements, curves, and unique geometries are locked into a third-party database, whether it belongs to a massive tech giant or an unknown app developer, it rarely disappears.
Enjoy the AI tools. Have fun experimenting with the vintage filters and the futuristic art generators. This technology is firmly here to stay, and it truly has incredible potential for human joy, efficiency, and creativity.
But, before you press upload on the very next viral trend, force yourself to take two minutes. Check what app you are actually using. Read how your images will be stored and who is allowed to access them. And most importantly, make sure that you understand exactly what your face might be worth to somebody else.
The writer is an electronic engineer with a background in information technology and sustainability
-----------------
The views and opinions expressed in this column are those of the writer, and do not necessarily reflect those of this publication