Artificial intelligence played a significant role in 52% of the advertising fraud schemes identified between January and September 2026, according to new research from ADEX, the AI-powered traffic validation and anti-fraud platform within AdTech Holding.
The findings come from ADEX’s State of Ad Fraud 2026: ADEX Trust & Safety Report. The report draws on aggregated moderation data from eight advertising platforms that use ADEX, covering 1.2 million campaigns and 4.9 million creatives.
Bots, Brand Clones and Deepfakes Lead AI Abuse
According to ADEX, bots impersonating humans appeared in 64% of AI-involved fraud schemes. Cloned brands and fake endorsements featured in 53%, and deepfakes appeared in 11%. Scam ads on large social platforms accounted for 10%.
The categories overlap. ADEX notes that a single fraud scheme often combines several AI-powered techniques at once. The percentages are calculated across distinct schemes rather than individual campaigns or accounts.
The report says AI has not invented new fraud. It has made existing fraud cheaper, faster and more convincing. Work that once required designers, copywriters, developers and translators can now be handled by a set of AI agents.
“The scheme itself has not changed in years: borrow a trusted face, promise big returns, move the victim into a private chat. What AI changed is the finish. The page looks real, the video looks real, and the person in the chat never sleeps and never makes a typo,” said Sergey Martianov, Chief Product Officer at ADEX.
Same-Day Comebacks After Rejection
ADEX researchers observed revised versions of rejected campaigns returning within hours, sometimes on the same day.
The company does not report a standard turnaround time because it varies widely between schemes. Even so, the trend turns moderation into what the report calls a speed contest.
According to the ADEX report, brand cloning and fake endorsements cluster in four areas:
- investment scams
- blockchain and crypto scams
- fraudulent e-commerce stores
- tech-support scareware that imitates well-known technology brands
Fewer Campaigns, Better Disguised
ADEX found that fewer policy-violating campaigns were submitted for moderation in 2026, but the ones that were submitted were more sophisticated.
Rejections fell by roughly a third after the first quarter. Operators moved away from high-volume submissions toward fewer, better-disguised campaigns.
“Fewer rejected campaigns do not mean a quieter market: bad actors are submitting fewer, better-built campaigns,” Martianov said.
AI Shopping Agents Could Be Next
Looking ahead, ADEX forecasts that in 2027 fraudsters will increasingly target the AI agents that browse websites, compare products and make purchases on users’ behalf.
The main concern is prompt injection, where malicious instructions are hidden in content an AI agent reads. Possible hiding places include invisible page elements, metadata, product images, or text addressed directly to AI assistants.
Attackers could also cloak pages, showing one version to humans and another to AI agents.
ADEX stresses that this is a forecast. Prompt injection does not yet appear as a separate category in its moderation data, although the techniques are well documented in security research.
“In 2027, the most valuable user on the web may not be a person at all. If an agent is doing the shopping, someone will try to scam the agent,” Martianov said.
Recommendations
ADEX advises advertisers to:
- monitor for their brand in other people’s ads
- verify where their clicks actually land
- ask technology partners how they distinguish legitimate AI agents from malicious automation
The full findings are available in the ADEX State of Ad Fraud 2026 report.