Sociable: Meta introduces new AI-powered tools at Advertising Week
The company is testing importing and creative insight features, as well as improved conversational guidance for its artificial intelligence elements.
Facebook and Instagram owner Meta disclosed on Wednesday that it took action on 5.3 million pieces of child sexual exploitation content on its platforms in India alone during the first half of 2026. The company reported that over 98% of the content on Facebook and Instagram in India was detected and dealt with before anyone reported it.
Globally, Meta acted on 33.2 million such items in the same period, with more than 97% found proactively. To strengthen its efforts against child sexual exploitation, Meta announced new AI tools to tackle adversarial tactics and enhance the effectiveness of systems for detecting ads leading to child sexual abuse material or directing users to such illicit activities off-platform.
The company emphasized that safeguarding young users remains a top priority and that predators have adapted their tactics to exploit ads, which can appear harmless but lead to illegal content hosted elsewhere. To counter this, Meta introduced a large language model system to detect "signposting," which involves identifying ad content that may seem benign but is strongly suspected of directing users to illegal material.
Additionally, Meta improved its assessment of the content an ad leads to, beyond just its displayed material, and integrated AI-driven sweeps to uncover material that earlier systems may have missed. The tech giant also employed a red-teaming AI agent to probe its own defenses for vulnerabilities and bolstered detection of banned users who create new accounts.
Meta continues to utilize PhotoDNA and other image and video matching technologies, which it has employed across its platforms since 2011, alongside behavioral signals and rules-based systems. The company shares hashes of identified material with other organizations via the Tech Coalition's Lantern program and blocks links to external sites hosting such content, deleting associated posts, comments, or ads.
Meta reported confirmed material to the US-based National Center for Missing & Exploited Children. In September, Meta committed to reporting child safety cases directly to India's National Cyber Crime Reporting Portal. The company's investigators, including former FBI personnel, focus on disrupting networks of predatory accounts rather than handling accounts individually.
Meta reiterated its commitment to collaborating with authorities to hold those behind these crimes accountable.
Written by urgent.news from YourStory's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.