Artificial intelligence in fitness: advertising and fake transformations
Artificial intelligence has become a regular feature in the online fitness world. In just a few years, social media feeds have been flooded with workout videos, weight loss programs, and muscle-building plans that promise radical changes in a very short time. In the middle of all this, one thing really stands out: the growing use of fully AI-generated characters and bodies, presented as if they were real people who supposedly followed that method and achieved almost magical results.
According to recent reports, including from BBC Sport, many of these ads show supposed before and after shots in just a few weeks, with promises like lose 40 pounds in one month, look 20 years younger, or change your entire body in 28 days. Health and exercise science experts point out that this kind of transformation, in such short time frames, is scientifically unlikely and, in many cases, simply impossible without serious risks to the body.
To make matters worse, platform dynamics help spread this kind of content. All it takes is for a user to watch a workout video or click on a diet ad for the algorithm to start pushing an endless sequence of similar material. In no time, the feed becomes a loop of aggressive promises, perfect bodies, and miracle programs. And more and more, a big part of this is made with generative AI, mixing synthetic avatars, fabricated testimonials, and heavily manipulated images, often without any warning that it is a simulation.
The result is a scenario that mixes advanced technology with a real regulatory Wild West, where the line between creative advertising and misleading content is getting thinner and thinner. And the person on the other side of the screen, trying to improve their health, lose weight, or gain conditioning, ends up exposed to unrealistic expectations, false hopes, and in many cases frustration and a direct hit to their self-esteem.
AI, social media, and the fitness content overload
The starting point for this phenomenon is the very nature of social media. Platforms like Instagram, TikTok, and others use AI models to quickly learn what grabs each person’s attention. If someone shows interest in workouts, diets, or body transformation, the system picks that up and starts prioritizing those kinds of videos. There is no simple off switch for that flow. As professor Andy Miah from the University of Salford explained, once this kind of content enters your radar, the trend is for it to multiply in your feed.
These algorithms are not concerned with safety, balance, or health: their priority is engagement. And what drives the most engagement? Shocking images, bold promises, and fast transformations. That is where artificial intelligence comes in strong, because it makes it possible to create exactly this kind of material cheaply, quickly, and at almost infinite scale.
With image and video generation tools, advertisers can produce characters that never existed, with sculpted bodies, flawless skin, and perfectly defined muscles. These avatars are placed in beach, gym, or home settings, with ready-made stories of overcoming adversity, claiming they followed a specific training plan and changed their lives in a month. But none of that is real. Even so, for viewers, it looks like genuine testimonials.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
Another important detail: the average user rarely manages to reliably tell whether those images are manipulated, real photos, or AI-generated. Some platforms claim to have systems to label synthetic content, but in practice many notices are hidden, shown in tiny text, or simply do not appear.
Unreal promises and scientifically unlikely transformations
A core element of this type of ad is in the explicit promises. Phrases like change your body in 4 weeks, look 20 years younger, or lose 40 lb in a month appear frequently in campaigns using AI avatars. Experts in exercise physiology, nutrition, and sports medicine are clear: results that extreme, in such short time frames, are not compatible with healthy processes for the vast majority of people.
Sustainable weight loss takes time, gradual diet adjustments, proper sleep, stress management, and a workout routine that matches each body’s reality. Gaining muscle mass also takes months of consistent stimulus, progressive overload, and proper recovery. When an ad promises something way outside that pattern, the chance it is misleading content is extremely high.
Professor Andy Miah describes the current scenario as a true lawless West in regulatory terms. According to him, the claims about how fast results can be achieved are completely unrealistic and feed false hopes, creating expectations that will almost never be met. This is not just a problem of exaggerated marketing: it directly affects the emotions of people who believe those promises.
What it looks like from the front lines of in-person training
Far from filters and virtual reality, professionals who have worked for decades with real people in real environments see this trend with huge concern. One example is instructor David Fairlamb, who has been working for over 30 years with group training and in-person coaching. For him, nothing replaces human contact, reading a student’s body, adapting exercises, encouraging face to face, and being responsible for closely monitoring progress.
When confronted with AI-generated ads that promise transformations in 28 days, his reaction is blunt: that simply does not happen. He says, with decades of experience to back him up, that there is no magic. Consistent change takes time, discipline, and especially proper guidance. Reducing everything to a before and after in a month is misleading and dangerous.
David’s daughter, Georgia Sybenga, who also works in the field, points out another sensitive issue: even people who grew up with social media struggle to tell what is real from what is digitally created. She herself admits that many times she has to look twice to figure out if she is seeing a real person or a very well made avatar.
They are both especially worried about younger users. Constant exposure to perfect bodies, often artificial, creates an unreal reference point. A teenager might look at a 30-day transformation, believe it is achievable, and when they do not get anywhere near that result, feel like a failure. Mental health is at risk, especially for boys and girls who are already insecure about their own bodies.
Georgia also warns about the physical risk of following totally generic AI-generated training plans that do not consider past injuries, mobility limitations, or specific health conditions. An exercise that is simple for one person may be dangerous for someone with knee, back, or cardiovascular problems. When a training program is sold as a universal solution without that kind of personalization, the chance of injury goes up.
The role of regulators and the challenge of spotting AI
In the middle of all this, regulatory bodies are trying to keep up. In the UK, the Advertising Standards Authority (ASA) is responsible for assessing whether ads are misleading or potentially harmful. The organization does not ban the use of artificial intelligence in campaigns, but it does look closely at the type of message being delivered.
Adam Davison, ASA’s director of data science, explains that the focus is not on the use of the technology itself, but on whether the ad misleads the public or can cause harm. He reports that in the past year the agency received around 300 complaints related to ads that in some way involved AI, and that number is growing.
One of the basic challenges is that sometimes not even analysts can say for sure whether an image or video was generated with AI. The tools are getting more sophisticated, avatars are increasingly similar to real people, and detection does not always keep up with that level.
Another issue is that AI makes it easy to quickly create campaigns by people or companies who are not familiar with advertising rules. We are not just talking about big brands, but also small business owners and affiliates using image generators, automated video editing, and ready-made text tools to launch ads for weight loss programs, supplements, or miracle workouts.
When ASA finds cases where the promises are too unlikely to be substantiated, the agency can step in. In many situations, especially when there is no record of previous violations, they send guidance notices, known as advice notices, explaining how the ads must be adjusted to comply with the rules and avoid misleading the public. The idea is to educate and steer the market before moving on to tougher measures.
Platforms, labels, and the struggle to avoid AI-generated content
In this game, social networks also share responsibility. Companies like Meta (which owns Facebook and Instagram) and TikTok say that artificial intelligence-generated content should be clearly labeled. The problem is that in practice many videos using AI to create bodies, faces, or environments do not display this kind of notice in a visible way.
Reports indicate that TikTok has already labeled more than 1.3 billion videos as AI-generated, while Meta says it analyzes this type of material using markers embedded in creation tools. But that is still far from covering everything that circulates on these platforms, especially when creators use external apps and services to generate images and then upload the content as if it were a regular video.
Many users interviewed in journalistic investigations say they would like to have the option to simply turn off AI-generated content, or at least filter it from their feeds. So far, however, platforms have not clearly indicated that they plan to offer that level of control to the public.
Meanwhile, the amount of AI-made material just keeps growing. As professor Andy Miah pointed out, the very logic of the attention economy fuels that growth: synthetic content is cheap, scalable, and can be tweaked in real time to maximize clicks, likes, and shares. From a reach and cost standpoint, it makes perfect sense for advertisers. From the perspective of those consuming it, the risks are much more complex.
When AI can help instead of deceive in the fitness world
Amid so much distortion, it is still possible to use artificial intelligence in a healthier, more honest way in the fitness space. The same technology that now powers misleading campaigns can support realistic training, track health indicators, and better guide beginners.
Apps can use AI to tailor workout suggestions to a person’s actual level, taking into account age, history, limitations, and realistic goals. Data analysis systems can show progress in strength, endurance, and well-being over months, instead of focusing on promises of 30-day change. Monitoring tools can help avoid overtraining by spotting signs of fatigue, performance drops, and even highlighting the importance of rest.
On the advertising side, AI models can also be configured to avoid certain practices, such as promises of extreme weight loss in very short time frames or the use of manipulated images without labels. Platforms can set clearer limits for what can be advertised when it comes to health, weight loss, and body transformation, requiring at least minimal proof of effectiveness for more aggressive methods.
At the same time, serious creators and professionals can use AI to improve educational content, make explanations more visual, show the real progress of clients over longer periods, and reinforce the idea that consistent progress takes time. This creates an important counterweight to the sea of misleading content and helps rebuild public trust in those who are truly committed to health, not impossible shortcuts.
In the end, the problem is not AI’s presence in the fitness world, but how it is being used in many campaigns: more as a machine for creating unrealistic expectations than as a support tool. The technology is already here, evolving fast, and it is not leaving social networks. The challenge is to make it stop being a tool for illusion and turn it into an ally for more conscious choices, more transparent communication, and a view of body and health that is a bit more human and a little less artificial.
