A study by Internet Matters examining the Online Safety Act's effectiveness has found that while the regulatory framework has increased visibility of platform-level safety features, significant gaps remain in protecting children from online harm. The research, which combined quantitative surveys with seven focus groups of children aged 11–16 and their parents, revealed that nearly half of all children continue to encounter harmful content despite the new regulations, with screen time management and AI-generated content risks insufficiently addressed by the current legislative approach.
The findings underscore a critical limitation of perimeter-based regulatory models: platform-native filters and age-verification gates, however well-intentioned, are routinely circumvented by young users and fail to address harms that occur within direct messaging channels. Real-time detection of threats as they emerge in private conversations—rather than reliance on blunt access restrictions—offers a more precise operational answer to the risks the Internet Matters study documents. Guardii monitors children's direct messages across Instagram, Snapchat, Discord, Roblox and other platforms, with detection modules for grooming, sextortion, child sexual abuse material (including AI-generated and deepfake variants), age-inappropriate contact, cyberbullying, and harassment. By identifying threat patterns without reading every message, the platform intercepts hostile contact before it reaches the child and surfaces evidence to parents, schools and law enforcement when escalation is warranted. While the Online Safety Act represents progress in platform accountability, the research confirms that targeted AI-driven intervention—of the kind Guardii, a Meta Business Partner backed by Startmate, delivers—remains indispensable to preventing the harms that legislative frameworks alone cannot intercept.
