A user reported a deceptive YouTube advert mimicking an iPhone storage warning. Despite multiple reports, Google initially approved the ad, citing policy compliance. The advert mimicked iOS system alerts with fake "Yes" and "No" buttons to coerce clicks.

Google's own Gemini model quickly flagged the ad for Misrepresentation and Deceptive UI Components. This highlights a discrepancy between automated detection and human review processes. The incident raises questions about Google's incentive structure, as high-click ads may generate revenue despite policy violations.

While AI can identify violations in seconds, the manual review appears slower or less effective. This case underscores the need for stronger integration of AI tools in ad screening to protect users from fear-based tactics and misleading designs that mimic operating system dialogs.