AI incidents in marketing

A practical archive of AI failures, backlash, and governance lessons in advertising and brand communication.

This site collects examples of AI incidents in marketing, especially AI advertising cases where automation, synthetic media, or generative creative tools created reputational, ethical, or brand-safety problems.

The Taylor Swift deepfake and Le Creuset's brand safety problem

Scammers used an AI-generated version of Taylor Swift to promote a fake Le Creuset giveaway.

Case note published on September 26th, 2026

Google pulls Gemini Olympics ad after backlash

Google draws criticism after an Olympics campaign to show Gemini helping a child write a fan letter to her sporting hero.

Case note published on September 26th, 2026

The chocolate factory that existed only in the prompt

When AI-generated marketing goes wrong.

Case note published on September 25th, 2026

Klarna's AI customer service course correction

Klarna course-corrects after its AI-first customer service strategy.

Case note published on September 25th, 2026

Duolingo’s AI-first announcement and the communications backlash

When an AI strategy statement contradicts a brand built on human warmth.

Case note published on September 21st, 2026

Show more Show less

Air Canada was held responsible after its customer service chatbot gave a passenger incorrect information about a bereavement fare refund.

When an AI chatbot’s mistake becomes the company’s legal responsibility.

Case note published on September 21st, 2026

The chatbot that wrote a poem about quitting

A software update disabled DPD's chatbot guardrails, allowing it to swear, criticize the company, and recommend competitors.

Case note published on August 25, 2026

French watchdog challenges Meta's job ad algorithm

France's equality watchdog ruled that Meta's Facebook job-ad delivery system produced discriminatory outcomes.

Case note published on January 9, 2025

McDonald’s Netherlands and the AI Christmas ad pullback

A case in tone, timing, and emotional mismatch.

Case note published on December 31, 2025

Guess, Vogue, and the AI-generated model debate

Beauty standards, disclosure, and the future of creative labour.

Case note published on December 31, 2025

About this archive