A growing AI marketing risk is catching business owners off guard, when synthetic images start looking like real people and spark backlash no one planned for.
There is a new kind of risk showing up in marketing right now.
It is not about copyright.
It is not exactly about deepfakes.
And it is not always about intent.
It is about what happens when AI generates a person who does not exist, yet looks close enough to someone real that people believe their identity has been used.
This is not a hypothetical scenario anymore. It is happening in real campaigns, in real businesses, with real consequences.
And if you are using AI in your marketing, it is something you need to understand.
Fictional people are no longer risk-free
For years, advertisers have used archetypes.
Stock photos, illustrations, models cast to represent a “type” of customer. That has always been normal.
AI changed the game because it can now create faces that feel authentic, nuanced, and highly specific. You can describe age, style, personality, fashion, lighting, emotion. The result is not a generic placeholder. It looks like someone you might know.
Here is the ethical tension.
A marketer may prompt an image of a confident entrepreneur, or a young fitness influencer, or an attractive woman with long brown hair and blue eyes. The AI is not trying to copy a real individual. But the output can land uncomfortably close to a recognizable person.
When that happens, intent does not always matter to the people who feel affected.
They see their identity reflected back at them in a commercial context they did not choose. That feels invasive. Even if technically the image is synthetic.
This is the new gray zone.
The difference between deepfakes and accidental resemblance
The word deepfake gets used too loosely.
Traditionally it means deliberately fabricating a real person’s likeness. Often with deception or harm in mind.
What we are seeing now is different.
AI can create what you might call accidental identity proximity.
No name. No source photo. No direct copying. Yet the resemblance is strong enough that audiences connect the dots.
This is not always malicious.
But it can still be damaging.
Reputation, dignity, and perceived endorsement are powerful forces in the digital economy. When an AI generated image appears to represent a specific person, especially in advertising, the emotional and commercial stakes rise quickly.
Commercial use raises the ethical bar
A synthetic image in a private brainstorming session is one thing.
A synthetic image in a paid campaign is another.
Advertising introduces money, influence, and implied trust.
It also introduces power dynamics. A brand using an AI generated face may have far more reach than the individual who feels they have been visually imitated.
That imbalance is where ethical responsibility lives.
It is not enough to say the image was fictional.
The real question is whether a reasonable observer could believe a real person’s likeness was being used or referenced.
If the answer is yes, the marketer now has a judgment call to make.
Realism plus sexualization equals higher risk
Another pattern is emerging.
The more realistic the image, and the more it is presented in a sexualized or emotionally charged context, the greater the potential backlash.
This is not just about sensitivity.
It is about perceived harm.
When a synthetic image that looks like a real individual is used to sell something in a suggestive or intimate way, the reaction can escalate fast. What may have started as a creative shortcut suddenly becomes a reputational crisis.
This is one reason lawmakers around the world are moving aggressively on nonconsensual synthetic imagery. The technology has outrun the norms.
Disclosure helps, but it is not a cure
Some marketers assume they can solve this by adding a line that says “AI generated image.”
Transparency is good. It builds trust.
But disclosure alone does not fix a likeness problem.
If the image still appears to represent a specific person, audiences will respond to the resemblance, not the fine print.
Responsible use requires more than labeling. It requires foresight.
A practical framework for responsible AI imagery
Businesses do not need to panic.
They do need a new checklist.
Before using AI generated people in marketing, consider five questions:
- Could this image reasonably be mistaken for a real, recognizable individual
- Does the campaign context imply endorsement, lifestyle, or behavior that could harm someone if misinterpreted
- Is the portrayal sexualized, controversial, or emotionally charged
- Are you prepared to replace the asset quickly if a credible concern is raised
- Would you feel comfortable defending the choice publicly as thoughtful and respectful
These questions are not legal advice. They are brand survival advice.
Why this matters for the future of marketing
AI is not just making content creation faster.
It is changing the relationship between identity and media.
We are entering a world where synthetic humans will appear in ads, training materials, websites, and social media at massive scale. That will create incredible opportunities for storytelling and personalization.
It will also create new ethical expectations.
Audiences are already signaling that they care about authenticity, consent, and fairness. Businesses that ignore this will learn the hard way that technological capability does not equal social acceptance.
The winners in the next phase of AI marketing will not be the ones who push the boundaries recklessly. They will be the ones who combine creativity with restraint.
The real lesson
The most important shift is philosophical.
The old marketing question was, “Can we do this?”
The new question is, “Should we do this, and how carefully?”
AI gives us tools that amplify imagination.
It also amplifies consequences.
Creating fictional people is not inherently unethical.
Using them without considering how real they may feel to others can be.
That is the learning moment for all of us.
If you are building campaigns with AI right now, take the extra minute to think through identity, perception, and impact. A small adjustment at the creative stage can prevent a very public problem later.
Technology moves fast.
Trust moves slowly.
The brands that remember that will have a much easier time navigating the future.

