Security1 publisher3 min readPublished Updated
Meta cleared 43 of 55 reported AI-CSAM ads in India as compliant with its Advertising Standards
The Tech Transparency Project counted 332 AI-generated child sexual abuse ads across five markets in ten months, and 78 of the 84 it found in India ran after New Delhi ordered Meta to clear such ads off Instagram.
The Watch · Security desk

What happened
- The Tech Transparency Project said on Tuesday it found 332 ads carrying child sexual abuse material running on Facebook and Instagram in the US, UK, EU, Australia and India over the past ten months.
- Eighty-four of those ads ran in India, and 78 of them were published after the Indian government ordered Meta to remove CSAM ads and met senior company officials about it.
- For 43 of the Indian ads that TTP managed to report, Meta answered that the ad does not go against its Advertising Standards.
- TTP says more than a dozen ads were still appearing on Meta platforms after it had made the company aware of its findings.
- TTP says the ads were built from photographs of real children, one of them a member of a European royal family, altered with AI.
Compiled by The WatchSomething wrong?How this is made
Why it matters
- constraint Because Meta accepts reports only on live ads, 29 of the 84 Indian ads had no reporting route at all, so any outside audit of the paid layer is capped at whatever is still running on the day the researcher looks.
- contradiction Meta's mitigation is reach and TTP's finding is acceptance, and if both accounts are accurate the same queue blocked and approved the same inventory; only Meta holds the delivery data that reconciles them.
- precedent A ruling for the petitioners would relocate the platform's CSAM reporting duty in India from a US clearinghouse to Indian police, and give other jurisdictions a template for demanding the same.
Paid inventory is the surface where Meta has the most to work with before anything reaches a user. Someone opens an account, pays, and submits a creative. Meta says it operates systems that block violating ads, with monitoring behind them for cases it initially misses [16], and its own policy bars ads that sexually exploit or endanger children [21]. The 332 ads the Tech Transparency Project counted were accepted for delivery, not caught by a takedown queue after publication [1].
The duplication looks like a control failure. TTP describes one ad in India, a photo of a preteen girl animated into a graphic sex act over a voiceover saying "there are no restrictions", posted by more than 50 different handles over a fortnight in August; the BBC says it has not seen that ad [11]. Matching fifty submissions of one creative over two weeks should be a straightforward case for any detection system. TTP says almost every ad in its set follows the same construction: a still photograph of a child, then AI animation into a sexual act [10].
The reporting figures are narrow, and they are the only external measurement anyone has. Meta offers the report option only while an ad is live, which left TTP able to file on 55 of the 84 Indian ads [12]. Of those 55, eleven came back as already removed and one is still unanswered [14]. Seventy-eight percent of what the researchers could put in front of the system was returned as compliant [22].
The timing lays out the sequence in the record. India's government ordered Meta to remove all ads and content promoting CSAM on Instagram roughly two months before the TTP report, following a BBC Eye investigation that found Instagram running paid CSAM ads in the country [7]. Six of the 84 Indian ads predate that order [24].
Meta's answer is about reach: many of the flagged ads had already been removed, most drew fewer than 200 impressions [16], and anyone on the platform can report an ad [26]. On child exploitation it reports to NCMEC, the US-based clearinghouse, in line with its legal obligations [17]. Bhuwan Ribhu of Just Rights for Children, a network of more than 250 organisations in India, says that is not enough, because Indian child protection law requires reporting to Indian law enforcement [18]. Distribution of CSAM is a criminal offence in India [21]. The National Commission for Protection of Child Rights and the National Human Rights Commission say they are examining the matter [20], and the Just Rights for Children petition asks the Supreme Court to direct platforms to report and to identify whoever promotes the material [19]. For a paid ad, that promoter is the party that bought the placement, and the buyer's details sit with the platform that took the money.
What to watch
- Whether Meta releases per-ad spend and delivery data for the 332 ads instead of the summary that most drew fewer than 200 impressions.
- Whether India's Supreme Court takes up the Just Rights for Children petition on mandatory reporting and identification of promoters.
- Whether the NCPCR or NHRC moves from examining the matter to an enforcement action under India's criminal provisions.