The Electronic Frontier Foundation has urged lawmakers to reject the KIDS Act, arguing that its combined rules for child safety, chatbots and explicit content would create practical pressure for online services to determine the ages of adults as well as minors. The group’s analysis arrived as Congress prepared to consider the package through an expedited process.
The legislation combines a revised Kids Online Safety Act, or KOSA, with several other bills and reporting requirements. KOSA includes language saying it should not be interpreted as requiring age verification. EFF argues that other provisions weaken that assurance because platforms must provide protections when they “know or should have known” a user was a child under 13 or a teenager aged 13 to 16.
That negligence-style standard could leave a service defending, after a complaint, whether it should have recognized a minor. EFF predicts that some operators will reduce the risk by asking for passports or driving licences, while others may use facial age estimation or infer age from account activity. Adults could consequently be asked to prove they are adults. This is the organization’s legal and practical assessment, not evidence that every covered platform will adopt the same method.
Other parts of the package add distinct thresholds. The SAFE BOTS Act would restrict certain chatbot features when a service knows or should know a user is a minor. The SCREEN Act would require services hosting sexually explicit material to decide whether a user is more likely than not below the applicable age. EFF says the differing schemes increase complexity, particularly for smaller companies with fewer resources for legal challenges.
The revised KOSA removed an earlier “duty of care” provision, which EFF described as a significant change. It would instead require covered platforms to establish and enforce policies addressing specified content and conduct. Some categories concern unlawful acts, while others include discussion involving alcohol, drugs, gambling and financial fraud. The group warns that platforms may over-remove lawful conversations about recovery, family problems or harm reduction to avoid liability.
EFF also raised concerns about direct messages, disappearing messages and encrypted services. Although the text contains a protection for strong encryption in some contexts, the group says that protection does not clearly resolve how platforms must address listed harms inside communications they cannot read.
The analysis identifies plausible incentives created by the bill but represents an advocacy organization’s interpretation. The key policy question is whether statutory disclaimers can prevent broad identity checks when compliance obligations still depend on knowing a user’s age, and whether narrower enforcement rules could protect minors without expanding data collection for everyone.
