Meta’s scam ad economy is out of control and fueled by AI deepfakes

Over 100 scam ads were detected on Meta platforms in a 24-hour period by the Financial Markets Authority regulators in New Zealand.

A single-day snapshot shows just how saturated platforms like Meta are with AI-generated deepfakes that convincingly mimic high-finance CEOs and politicians.

The fake videos and images make it appear as if trusted figures are endorsing investments, with brand logos such as Australian financial services firm Westpac and New Zealand-based Kiwibank adding to the deception.

Victims initially funnel into $250 (USD) “starter” deposits, with the small entry payments lowering suspicion and increasing conversion rates for scammers.

Over the course of a year, reported fraud losses hit $265 million (NZD) annually, according to B2B news. However, actual losses are likely nearer to $1 billion, given that 80% of scams go unreported.

The scammers operate as paying ad customers, adding legitimacy to what the end-user sees in their feed. It also means that Meta is unable to remove the fraudulent ads until it's too late.

In this shady economy, new scam ads appear faster than old ones can be taken down, with content removal lagging behind ad creation speed. Nothing sits around for very long. Nothing is stagnant.

Many individuals and businesses in the financial, tech, and property sectors are targeted by deepfake scams, with templates easily reused across industries and scaled up at an industrial pace.

Online safety non-profit organization Netsafe found that a sizable 72% of adults in New Zealand are exposed to scams annually, with most encountering scam content regularly, even if they don’t engage.

Nearly one in four exposed users end up financially impacted, with the average loss sitting at $3,352 (NZD – just short of $2000 USD.) Ominously, over 40% of victims recover nothing at all after reporting the crime.