Hey all, this is a quick update on AI news and events.
On March 20, 2026 the Trump administration released its long-awaited legislative recommendations for a national AI policy framework. The proposals, spearheaded by White House Special Advisor David Sacks and OSTP Director Michael Kratsios, touch on children’s online safety, state-law preemption, intellectual property, AI literacy, and more. For privacy and compliance professionals, several elements deserve close attention.
Children’s Safety Takes Top
The framework leads with children’s protections consistent with the administration’s posture since its December 2025 executive order on AI. Key proposals include reaffirming that existing child privacy laws apply to AI systems (including limits on data collection for model training and targeted advertising), mandating commercially reasonable, privacy-protective age-assurance mechanisms such as parental attestation, and equipping parents with tools to manage screen time, content exposure, and account controls. The repeated use of the phrase “likely to be accessed by minors” signals the administration may be open to the acknowledged standard that has been a flash point in debates over the Kids Online Safety Act and COPPA 2.0.
Preemptions with Carve-Outs
The framework calls broadly for federal preemption of state AI laws that could hinder innovation, but it carves out notable exceptions. States would retain authority to enforce generally applicable laws protecting children, exercise traditional police powers, and govern public-sector AI procurement. Beyond those carve-outs, the language stays high-level: there are no direct references to state ADMT or AI-chatbot rules, though the recommendation that states be barred from regulating AI development altogether and from holding developers liable for third-party misuse will be closely watched.
A Light-Touch Enforcement Vision
The administration explicitly opposes the creation of a standalone AI regulatory body. Instead, it favors sector-specific oversight through existing agencies supplemented by industry-led standards. Regulatory sandboxes also feature prominently; Senator Ted Cruz already has a bill proposing a two-year sandbox program with exemptions from federal rules that impede product development.
What Practitioners Should Watch
These recommendations are advisory. Congress still has to legislate. But the framework establishes the administration’s negotiating position and aligns closely with Senator Marsha Blackburn’s companion AI framework, released just one day earlier. For compliance teams, the key takeaways are: (1) children’s privacy enforcement around AI is accelerating at the federal level, (2) the preemption debate is live, and the contours are still being drawn, and (3) the preference for existing-agency oversight and sandbox models means the regulatory landscape will remain fragmented for the foreseeable future.
Stay tuned, this is going to move fast!
CIPAWorld 