The recent loss of a legend like Piyush Pandey feels like a marker between two eras of advertising: one defined by instinctive human storytelling, and the emerging one defined by machine-generated imagination.
Today, AI-generated advertising is advancing faster than the industry’s ability to evaluate it critically. Over the past year, the gap between ‘AI that accelerates creativity’ and ‘AI that exposes creative shortcuts’ has become impossible to ignore. Some campaigns are rewriting production efficiency benchmarks; others are reminders that no amount of computation can compensate for the absence of judgment.
For the creative industry, this is no longer a conversation about technology adoption. It’s a conversation about craft.
The uncanny valley showing up across AI-led ads isn’t a glitch; it’s a creative integrity issue. [The uncanny valley refers to the eerie, unsettling feeling people get when something looks almost, but not quite, human, like a realistic robot or CGI character.]
Where emotional intelligence fails, AI shows its seams
Generative models can produce polished frames. What they cannot create is emotional logic, and the last year has made this painfully clear.
South Korea’s GS25 convenience store released an AI-generated campaign with distorted hands, stiff facial expressions and eerily symmetrical smiles. Public reaction was instant: “creepy,” “soulless,” “plastic.” The work became a case study in what happens when a brand outsources emotional nuance to a generator.
Coca-Cola faced similar backlash when its 2024 ‘Holidays Are Coming’ AI ad went viral across the US. Viewers called it “deeply derivative,” accusing the brand of diluting its warm, human-driven legacy with a synthetic imitation. Comments included calls for a boycott, with many saying the ad felt emotionally empty compared to their iconic holiday films.
In contrast, Nike’s ‘Never Done Evolving’ campaign with Serena Williams demonstrated what thoughtful AI integration can look like.
Through the use of real archival match data to simulate a game between a young Serena and her present-day self, the campaign earned applause for narrative depth, emotional clarity and innovation without gimmickry.
That is the difference between automation and authorship.
Cultural context is non-negotiable
Emotional logic was the first test. Cultural grounding is now the second.
The AI Barbie wave that swept through India earlier this year proved how easily cultural tone can slip. Dozens of brands generated Barbie-style posters featuring identical oval faces, Western beauty cues and a visual grammar that felt imported rather than local. The content went viral, but mostly as a parody. Audiences called the executions “algorithmic,” “same-looking” and “contextless.”
A more sensitive example emerged during the election season. Several AI-generated political videos attempted aspirational storytelling but were immediately mocked online for jittery faces, inconsistent lighting and emotionally flat expressions. Viewers tagged them as “unreal” and “unrelatable,” highlighting how quickly cultural intuition collapses when models imitate behaviour instead of reflecting lived experience.
Yet there were clear wins too, especially when AI was used to expand cultural insight. Nutella’s ‘Nutella Unica’ campaign created seven million unique jar designs using AI, but each design drew from real environmental patterns and artistic cues. The result? All seven million jars sold out within a month.
Closer home, Cadbury’s AI-assisted ‘Shah Rukh Khan-My-Ad’ resurfaced as a UGC (user-generated content) trend during Diwali. Local retailers jumped on the opportunity to create personalised Diwali ads featuring SRK, promoting their stores. The campaign was celebrated again for hyperlocal relevance, emotional warmth and an experience that felt authentically Indian.
In an industry obsessed with personalisation, the next competitive advantage is contextual precision.
Why some AI ads work and others don’t
Across markets, campaigns that succeeded showed two consistent traits: human-led emotional spine (the narrative and emotional logic came from people, not machines) and cultural clarity. The work understood its audience, their behaviour, humour, nuance and expectations.
On the flip side, campaigns that over-relied on model output, had unrealistic visual optimisations or lacked cultural grounding struggled to land with audiences.
The learning: audiences don’t reject AI; they reject AI used without taste.
Transparency isn’t optional anymore
Audiences have caught up. They can spot AI artefacts as easily as they once spotted stock photos, and their expectation has shifted from ‘don’t use AI’ to ‘don’t pretend you didn’t’.
Several beauty and apparel brands were called out this year for slipping in AI-retouched models without proper disclosure. The backlash wasn’t about the use of AI; it was about the secrecy. As one viral comment put it: “If you’re hiding it, it’s a shortcut, not creativity.”
Brands that embraced transparency did much better. Heinz openly shared its AI-experimentation process to show how AI interprets the brand’s visual heritage. BMW released behind-the-scenes content explaining its AI-assisted creative workflow.
In both cases, transparency reframed AI as craftsmanship, rather than cost-cutting.
When AI stops feeling uncanny
AI is not a threat to creativity. It is only exposing fragility in workflows built on speed rather than consideration. The uncanny valley is only a symptom of a deeper misalignment; when brands outsource emotions, the audience notices instantly.
However, when AI is paired with strong human insight, cultural clarity and transparent intent, it stops feeling uncanny and starts feeling inevitable.
This is the creative moment we’re in. Not man versus machine, but meaning versus automation.
Reena Rose is creative director at Wunderkint
