Table of Contents
How Countries Are Regulating AI: What Parents Should Know
The EU, US, and China are taking very different approaches to AI regulation. Here's what the EU AI Act, US executive orders, and global AI governance mean for families and kids.
If your kid wants to understand power in the 21st century, the best single topic to follow isn’t social media regulation, or climate policy, or even nuclear arms control. It’s AI regulation. What governments decide about which AI applications are permitted, who is liable when AI makes harmful decisions, and how AI systems should be disclosed to users will shape more of their adult lives than almost any other policy question.
Most parents don’t know these conversations are happening, let alone what’s in them. Here’s the essential map.
Key Takeaways
- The EU AI Act (2024) is the world’s first comprehensive AI regulation framework — it uses a risk-tiered approach, banning some applications outright and requiring conformity assessments for high-risk uses.
- The US has no comprehensive federal AI law; instead, multiple executive orders, NIST frameworks, and sector-specific rules are creating a patchwork.
- China has enacted several AI regulations focused on algorithmic recommendations, deepfakes, and generative AI — all requiring political alignment with government goals.
- Children are specifically mentioned in AI regulation in the EU AI Act (social scoring bans) and in proposed US legislation (Kids Online Safety Act, proposed COPPA expansions for AI).
- Raising a kid who understands AI governance as a civic issue — not just a tech issue — is a specific, teachable goal.
The EU AI Act: The World’s First Comprehensive AI Law
The European Union passed the AI Act in March 2024, with implementation beginning 2025–2026 depending on risk tier. It’s the most ambitious AI regulation in the world and will affect any company doing business in Europe — which includes virtually every major US tech company.
The framework is risk-based:
Unacceptable risk (prohibited):
- Social scoring systems by governments — ranking citizens based on their social behavior, as China’s social credit system does.
- Real-time remote biometric surveillance (facial recognition) in public spaces, with narrow law enforcement exceptions.
- AI manipulation using subliminal techniques that bypass rational agency.
- AI that exploits vulnerabilities of specific groups (children, people with disabilities).
- Emotion recognition in workplaces and educational institutions.
- Predictive policing based on profiling.
High risk (allowed but regulated):
- AI in hiring and employment decisions.
- AI in credit scoring.
- AI in criminal justice risk assessment.
- AI in medical devices.
- AI in educational assessment (grading systems, student evaluation).
High-risk systems require technical documentation, human oversight mechanisms, and registration in an EU database. The developers must conduct conformity assessments — essentially, demonstrate the system is safe and fair before deployment.
Limited/minimal risk: Most consumer AI applications — chatbots, recommendation systems, spam filters. Subject to transparency requirements (AI must disclose it’s AI) but minimal other regulation.
General-purpose AI (GPAI): Large foundation models like GPT-4 or Claude are subject to transparency rules about training data and are required to implement copyright compliance mechanisms. “Systemic risk” GPAI (most powerful models) face additional requirements including adversarial testing.
The US Approach: Executive Orders and Patchwork
The US has not passed comprehensive federal AI legislation as of 2026. What exists instead:
Executive Order 14110 (October 2023): The Biden administration’s executive order on AI required agencies to assess AI risks in their domains, directed NIST to develop AI safety standards, and required developers of the most powerful AI systems to share safety testing results with the government before deployment (under the Defense Production Act).
NIST AI Risk Management Framework (2023): The National Institute of Standards and Technology published a voluntary framework for managing AI risks — not a law, but widely adopted as a de facto standard by companies seeking to demonstrate responsible AI practices.
Sector-specific rules: The FDA regulates AI/ML medical devices. The FTC has taken enforcement actions against discriminatory algorithms. The CFPB oversees AI in lending. The EEOC has issued guidance on AI in hiring. The result is overlapping but incomplete coverage — some domains (healthcare, lending) have more protection than others (social media, education, entertainment).
The partisan dimension: Comprehensive AI regulation has stalled in Congress partly because of genuine disagreement about whether national AI competitiveness would be undermined by heavy regulation. Tech industry lobbying has also been substantial. The Trump administration’s 2025 executive order on AI rescinded the Biden-era order and directed a lighter-touch approach.
China’s Approach: State-Aligned AI
China has enacted several specific AI regulations:
-
Algorithmic Recommendation Regulation (2022): Requires transparency in how recommendation algorithms work, allows users to opt out of personalization, and mandates that recommendation algorithms cannot be used to endanger national security or “socialist core values.”
-
Deep Synthesis Regulation (2022): Covers deepfakes and AI-generated content — requires watermarking and disclosure, prohibits fake news generation, prohibits impersonating leaders.
-
Generative AI Regulation (2023): Requires that AI-generated content conform to “socialist core values” and government policies. AI companies must submit content generation models for government security review before deployment.
The pattern is different from the EU’s approach — Chinese regulation is less about protecting individuals from AI harms and more about ensuring AI supports rather than threatens state authority. This distinction matters for understanding the global AI governance landscape: there isn’t a single model of “responsible AI” — there are competing governance philosophies with very different values embedded.
AI Regulation That Specifically Affects Children
The children’s dimension of AI regulation deserves specific attention.
| Regulation | Jurisdiction | What It Covers for Children |
|---|---|---|
| EU AI Act | EU | Bans emotion recognition in schools; bans AI exploitation of minors; prohibits subliminal manipulation of children |
| COPPA (existing) | US | Children under 13 — online data collection rules apply to AI companion apps, tutoring AI, games |
| Kids Online Safety Act (proposed) | US | Would require AI platforms to minimize harm to minors; duty of care standard |
| UK Online Safety Act (2023) | UK | Age verification, default safe settings for services likely accessed by children |
| EU Digital Services Act | EU | Platforms cannot serve targeted advertising to minors based on profiling |
The UK Online Safety Act (2023) is the most comprehensive currently in force specifically addressing AI and children — it requires large platforms to conduct Children’s Risk Assessments and implement safety measures by design for users who might be minors.
For US families, the patchwork is genuinely thin. COPPA protects data collection from children under 13 but does not cover many AI behaviors (recommendation systems, manipulation, AI companions) that affect kids. Comprehensive federal children’s AI protection legislation has been proposed but not passed.
How to Teach Your Kid About AI Governance
Ages 5–8: Rules are made by people
Start with the basic concept: AI tools have rules about what they can and can’t do. Who made those rules? (People — engineers, lawyers, governments.) Can the rules change? (Yes.) Are the rules the same everywhere? (No.) This is civics at the simplest level — the world has rules, the rules are made by people, people can change them.
Ages 9–12: Compare the approaches
Pick one AI application — facial recognition in schools, for example — and look at how the EU, US, and China handle it differently. The EU AI Act prohibits emotion recognition in educational settings. The US has no equivalent prohibition. Some Chinese provinces have deployed classroom emotion monitoring. Ask: “Which approach would you want to live under? What values are embedded in each choice?”
Ages 13+: Read one regulation directly
The NIST AI Risk Management Framework (available free at nist.gov) is designed to be readable by non-specialists and explains risk categories, impact assessment, and governance structures. A teenager can read the “Govern” function chapter and understand the core framework. Connecting this to AI ethics and accountability makes the abstract governance discussion concrete.
The question to ask: “If you were writing the rules for AI in schools — what would you allow? What would you ban? Who would you make responsible if the AI caused harm?”
What to Watch For Over the Next 3 Months
Month 1: The EU AI Act’s implementation timeline runs 2024–2027 in phases. Follow the EU’s AI Office (digital-strategy.ec.europa.eu) for updates on high-risk AI provisions taking effect. These will affect AI tools used in schools and healthcare.
Month 2: The US Congress periodically considers AI-specific legislation. Track the progress of any kids-specific AI bills through congress.gov — the Kids Online Safety Act has had multiple versions. Notice what passes and what doesn’t.
Month 3: Your school district likely uses AI tools of some kind. Ask them: “What AI systems do you use? Are they assessed for bias? Who is responsible if the AI makes an error that affects a student?” These are the questions the EU AI Act requires high-risk AI users to answer. Asking them in a US school is civics in practice.
Frequently Asked Questions
Does the EU AI Act apply to US companies?
Yes, if they serve EU users. The same extraterritorial logic as GDPR applies — any company that markets to or has users in the EU must comply, regardless of where the company is headquartered. This is why the EU AI Act effectively sets global standards, similar to what GDPR did for data protection.
What happens if an AI company violates the EU AI Act?
Fines of up to €35 million or 7% of global annual turnover, whichever is higher — for prohibited practices. Fines of up to €15 million or 3% of turnover for other infringements. These are proportionally similar to GDPR fines, which have been levied against major US tech companies.
Is AI emotion detection in schools legal in the US?
As of 2026, there is no federal law specifically prohibiting it. Some states have student biometric data protection laws that might apply (Illinois’ BIPA, for example). The EU AI Act explicitly prohibits emotion recognition in educational institutions. US parents who have concerns about AI monitoring tools in their children’s schools should review their district’s acceptable use policy and state education privacy laws.
Can my child opt out of AI decision-making that affects them in school?
This depends entirely on the state and district. There is no federal right for students to opt out of AI-based grading, assessment, or risk scoring. Some districts have processes for human review of AI recommendations. The practical step is asking your school district what AI tools they use and what the appeals process is when an AI system affects a student’s placement, discipline, or services.
About the author Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.
Sources
- European Parliament and Council. (2024). “Regulation (EU) 2024/1689: The AI Act.” https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32024R1689
- National Institute of Standards and Technology. (2023). “AI Risk Management Framework (AI RMF 1.0).” https://doi.org/10.6028/NIST.AI.100-1
- The White House. (2023). “Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence.” https://www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence/
- UK Government. (2023). “Online Safety Act 2023.” https://www.legislation.gov.uk/ukpga/2023/50/contents/enacted
- Cyberspace Administration of China. (2023). “Interim Measures for the Management of Generative Artificial Intelligence Services.” https://www.cac.gov.cn/2023-07/13/c_1690898327029107.htm
- Federal Trade Commission. (2023). “FTC Report on Commercial Surveillance and Data Security.” https://www.ftc.gov/legal-library/browse/reports/ftc-report-commercial-surveillance-data-security
- Congressional Research Service. (2024). “Artificial Intelligence: Overview, Recent Advances, and Considerations for the 118th Congress.” https://crsreports.congress.gov/product/pdf/R/R47433
