The Trump administration is reportedly considering an executive order to establish federal oversight of new AI models, marking a potential reversal from campaign promises to reduce AI regulation. According to WIRED’s Uncanny Valley podcast, the administration is exploring federal oversight mechanisms that would represent a significant shift in approach.
This development comes as industry leaders increasingly call for AI regulation. Billionaire hedge fund manager Paul Tudor Jones told CNBC that the U.S. is “late to regulating AI” and needs immediate action on watermarking to distinguish deepfakes. Jones noted that 80% of participants at a recent AI conference supported regulation, up from just 20% last year.
Growing Industry Support for AI Oversight
The shift in sentiment among AI experts reflects mounting concerns about unregulated AI deployment. Jones emphasized the urgency of implementing watermarking systems to combat deepfakes, which have become increasingly sophisticated and harder to detect.
The March 2026 White House AI policy framework provided initial guidance, but industry leaders argue more comprehensive federal oversight is needed. The framework addressed basic safety measures but stopped short of mandatory compliance requirements for AI model developers.
Current regulatory gaps have become apparent in consumer markets, particularly with AI-powered children’s toys flooding the market without adequate safety controls.
AI Toys Highlight Regulatory Gaps
The unregulated AI toy market exemplifies broader oversight challenges. WIRED reported that AI toys marketed to children as young as three operate with minimal safety guardrails. By October 2025, over 1,500 AI toy companies were registered in China alone.
Testing by consumer advocacy groups revealed serious safety issues:
- FoloToy’s Kumma bear (powered by OpenAI’s GPT-4o) provided instructions on lighting matches and finding knives
- Alilo’s Smart AI bunny discussed inappropriate adult content including “leather floggers” and “impact play”
- Miriat’s Miiloo toy promoted Chinese Communist Party talking points during NBC News testing
Huawei’s Smart HanHan plush toy sold 10,000 units in China within its first week, while Miko claims over 700,000 units sold globally. The rapid market growth occurs without sector-specific safety standards.
Consumer Protection Concerns
R.J. Cross, director of the Public Interest Research Group, argues that current AI toy regulations are insufficient to protect children. The toys’ ability to engage in open-ended conversations creates unpredictable risks that traditional toy safety standards don’t address.
Consumer groups advocate for mandatory content filtering, age-appropriate response training, and transparent data collection practices for AI toys. Current voluntary industry standards vary widely between manufacturers.
Cybersecurity Lessons Shape AI Regulation
The evolution of cybersecurity regulation over the past two decades offers insights for AI oversight development. Dark Reading’s 20-year retrospective traces how cyber threats evolved from simple viruses to “industrial-grade operations that can disrupt hospitals, utilities, and supply chains.”
Key regulatory milestones in cybersecurity include:
- Mandatory breach disclosure rules following major incidents
- Critical infrastructure directives after attacks on utilities
- Sector-specific compliance obligations for healthcare, finance, and government
- Board-level accountability requirements for public companies
The cybersecurity regulatory framework took years to develop, often in response to major incidents. AI regulation advocates argue that proactive measures could prevent similar reactive policymaking.
Federal vs. State AI Regulation Approaches
While federal oversight remains under consideration, state-level initiatives continue advancing. California’s SB 1001 requires AI chatbots to disclose their artificial nature, while New York City implemented AI hiring audit requirements for employers.
The European Union’s AI Act, which took effect in 2024, provides a comprehensive regulatory framework that many U.S. policymakers reference as a potential model. The EU approach includes:
- Risk-based classification for AI systems
- Prohibited AI practices including social scoring
- High-risk AI requirements for safety-critical applications
- Transparency obligations for general-purpose AI models
U.S. companies operating globally must already comply with EU AI Act requirements, creating de facto international standards.
What This Means
The reported Trump administration consideration of federal AI oversight represents a pragmatic response to mounting industry and security concerns, despite campaign rhetoric favoring deregulation. The 80% support rate among AI experts signals broad industry recognition that voluntary self-regulation is insufficient.
The AI toy market’s safety failures demonstrate how rapidly AI applications can outpace regulatory frameworks. Unlike traditional software, AI systems’ unpredictable outputs create novel liability and safety challenges that existing consumer protection laws don’t adequately address.
Federal AI oversight, if implemented, would likely focus on high-risk applications first — similar to how cybersecurity regulations prioritized critical infrastructure. Consumer applications like AI toys may require separate regulatory approaches given their direct interaction with vulnerable populations.
The timing suggests that AI regulation is moving from a partisan political issue to a bipartisan governance challenge, with both industry leaders and policymakers recognizing the need for proactive oversight before major incidents force reactive measures.
FAQ
What specific AI oversight is the Trump administration considering?
According to reports, the administration is exploring an executive order that would establish federal oversight of new AI models, though specific details about scope and enforcement mechanisms haven’t been disclosed publicly.
Why are AI experts now supporting regulation when they previously opposed it?
Paul Tudor Jones noted that support among AI conference participants jumped from 20% to 80% in one year, likely reflecting growing awareness of safety risks and potential misuse as AI capabilities have advanced rapidly.
How do current AI toys bypass safety regulations?
AI toys operate in a regulatory gap between traditional toy safety standards (which focus on physical hazards) and software regulations (which don’t address conversational AI risks). Most current oversight relies on voluntary industry standards rather than mandatory compliance requirements.
