Protecting Children from Chatbots Act

Scripture charges parents with the formation and protection of their children: "Train up a child in the way he should go" (Proverbs 22:6). That responsibility does not pause when a child picks up a phone. AI chatbots present a genuinely new category of risk, capable of forming persistent, emotionally manipulative relationships with minors, exposing them to harmful content, and operating entirely outside a parent's awareness. South Carolina's Protecting Children from Chatbots Act gives parents a legal foothold in territory that has, until now, been governed solely by the commercial interests of platform operators.

The bill's core mechanism is straightforward and sound. Covered entities must make a limited-access mode available and verify user age before granting access to full chatbot features. If a parent provides consent, a minor may use limited-access mode or, with explicit authorization, access restricted features. This structure does not ban children from technology. It restores the parent to the role the law has long recognized in other contexts involving minors, ensuring that meaningful gatekeeping exists and that it rests with the family rather than the platform.

Equally important is the bill's prohibition on prioritizing engagement at the expense of user well-being. The business model of many AI platforms depends on maximizing time-on-platform, and minors are particularly vulnerable to the psychological techniques used to sustain that engagement. Codifying a duty of well-being over engagement is not a minor regulatory footnote. It is a direct check on a documented and serious harm, and it reflects the principle that children are not simply revenue units whose attention is available for capture.

The bill also establishes procedures for reporting incidents of harm that a chatbot inflicts on a minor and provides for penalties against non-compliant entities. Accountability mechanisms matter. A law without enforcement is a suggestion, and the inclusion of both a reporting pathway and meaningful penalties signals that the legislature intends this protection to function in practice, not merely on paper.

The American Council supports this legislation as a concrete, enforceable expression of parental authority in the digital age. It does not overreach. It does not prohibit the technology. It requires that those who deploy AI systems toward children bear responsibility for how those systems are accessed and what they are permitted to do. We urge its passage and commend it as a model for other states navigating the same urgent challenge.