
BRAND CONTENT · AI
40-60%
CAN'T SPOT FAKES
70%
SHOOT COST · AVOIDED
$20-70K
UK DISCLOSURE RULES
0
An investigation by the Guardian has surfaced a practice that is quietly reshaping digital advertising: brands are commissioning AI-generated influencers — hyperrealistic synthetic humans — to produce what appears to be authentic user-generated content (UGC), with no disclosure requirement and creators bound by non-disclosure agreements. The practice spans apps, fashion, and design tools, and the regulatory vacuum is nearly total.
Extraction Map
The cases are specific. Once, a disposable-camera photo app, ran Instagram videos showing a bride crying over her wedding experience with the product — a bride that Reality Defenders, a deepfake detection firm, assessed as likely AI-generated. Maket, an AI housing-design app, confirmed using synthetic influencers as a "test" of creative concepts. Ashle, a Dubai-based fashion brand, deleted photos from its social media page after the Guardian inquired about an image featuring a woman with an extra finger — a telltale artifact of AI image generation. The economics are blunt. Clarissa Mansbridge, a former celebrity manager who now creates AI influencers under her Mia Metaverse portfolio, says brands are fleeing traditional photoshoots that cost $20,000 to $70,000 per session. Her estimate: 40% to 60% of content from major brands is now AI-generated, with creators silenced by NDAs. "I call it plausible deniability," she said. The regulatory landscape is a void. The UK's Advertising Standards Authority confirmed there is nothing in its rules prohibiting undisclosed AI-generated promotional content. The ASA's test is whether an ad is "misleading" about the product, not whether the person endorsing it exists. The EU's AI Act will require labelling of deepfake content starting August 2025, but this will not apply in the UK. Which? found that 70% of consumers cannot reliably distinguish real from AI-generated video. The extraction pattern is layered. Brands extract consumer trust by simulating authentic experiences that never occurred. Human influencers and photographers lose income to synthetic replacements that require no hourly rate, hold no opinions, and generate no bad press. Consumers bear the cost of degraded information quality — they cannot evaluate product claims when the "reviewer" is fictional. The NDAs ensure the deception remains structurally invisible. Mansbridge frames this as efficiency: "The authenticity of UGC was always about resonance, not who made it." This is the argument that the signal (authentic customer experience) can be separated from its source (an actual customer). It cannot. The entire value of UGC to consumers rests on the assumption that a real person chose to use and endorse a product. Synthetic UGC is not a more efficient version of the same thing — it is a simulation of trust that destroys the information value it exploits. Leeds-based artist Zac Rossiter was directly approached by a marketing agency offering to create AI-generated unboxing videos of his artwork prints. He declined. The offer illustrates how the practice is being actively marketed downstream — not just to large brands but to small businesses and individual creators, normalizing synthetic testimonials across the entire commerce stack.
Worth reading?
SKIP
Score
24
/100
Extractive — brands capture trust-value built by real human UGC creators, replace them with zero-cost synthetic equivalents, and use NDAs to prevent the market from pricing in the deception.
What Is GRIN?
GRIN stands for Generativity, Resilience, Innovation efficiency, and Novelty openness. These four dimensions capture the essential characteristics of any memetic intelligence — whether a nation, a corporation, a religion, or a movement.
About GRINGenerativity
1/5
Resilience
1/5
Efficiency
2/5
Novelty
1/5
Extractive
4/5
Inverted
Does this create new capability or just redistribute existing value?
Destroys more than it creates. The efficiency gains accrue to brands; the trust infrastructure that makes UGC valuable is consumed without replacement. Net value destruction.
Verified Facts (7)
70% of people are unable to correctly identify all the real and fake videos Which? showed them
Brands want high-end photography, but they don't want to pay $20,000 to $70,000 for a traditional photoshoot
About 40% to 60% of the content out there from some of the big brands is actually being made through AI, but a lot of the creators are under NDA
There's nothing in our rules that prohibits this and there are no disclosure rules for AI content labelling — ASA spokesperson
In the EU, new rules under the Artificial Intelligence Act will begin applying in August, requiring AI-generated or manipulated content such as deepfake images, audio and video to be clearly labelled. The legislation will not apply in the UK.
Reality Defenders, a cybersecurity company specialising in deepfake detection, assessed the Once brand has likely used AI-generated influencers
After being approached by the Guardian, Ashle deleted photographs from its social media page
More latest news
Read all