Meta Erases 4 Million Accounts Over Instagram Ad Scandal

Meta has pushed back against reports of exploitative Instagram ads, revealing that its automated AI safety systems have purged over 4 million suspicious accounts.

The social media giant is facing intense scrutiny following a wave of disturbing reports regarding its advertising systems, forcing the tech giant to defend its safety protocols publicly. Recent media investigations revealed that advertisements violating policies against child exploitation had managed to slip through the company’s automated safety nets on Instagram, specifically within India.

In an official response, Meta explicitly addressed the reports, stating that the company takes these concerns seriously, never wants this content on its platforms, and is committed to improving its efforts to combat it.

The company confirmed that its internal enforcement systems had already identified and disabled several of the violating advertisements and the accounts behind them before the public reports even emerged. Following a deeper internal investigation, Meta wiped out additional ad campaigns, banned associated business accounts, and blocked external web addresses connected to the malicious content.

The core of the recent controversy involved how these horrific ads were served to users. Meta strongly pushed back against any suggestions that its algorithms were designed to connect these bad actors with sensitive audiences.

The company stated that it is categorically inaccurate to suggest that they would knowingly and deliberately target ads featuring children to people based on an inappropriate interest in children. Instead, the company explained that its algorithm is tuned to do the exact opposite by tracking down and isolating accounts that show suspicious patterns.

Meta revealed that it used these automated detection tools to wipe out more than 4 million suspicious accounts globally across Facebook and Instagram last year alone. In India, where the recent ad scandal took place, the company’s focus on off-platform link coordination resulted in the removal of 160,000 predatory accounts over the last six months.