L1
·
Quiz
·
Lab
L2
·
Quiz
·
Lab
L3
·
Quiz
·
Lab
L4
·
Quiz
·
Lab
Module Test
Module 3 · Lesson 1

The US Approach

Voluntary frameworks, principles-based guidance, and the theory behind American AI governance

The NIST AI Risk Management Framework was published to minimal fanfare outside policy circles. No mandatory compliance deadlines. No enforcement agency. No fines.

And yet within months, major corporations were rewriting their AI governance policies to align with its structure. Voluntary can still be powerful — if the vocabulary becomes universal.

The American Philosophy

While the EU chose binding law, the United States historically chose a different path: voluntary frameworks, sector-specific guidance, and the principle that innovation should not be blocked by preemptive regulation. This isn't the absence of governance — it's a different governance theory. The underlying bet: that industry, if given clear frameworks and incentives, will self-regulate more effectively and adaptively than legislation can.

Whether that bet is paying off is the central debate in US AI policy.

The NIST AI Risk Management Framework

Published in January 2023, the NIST AI RMF is the primary federal voluntary framework for managing AI risk. It organizes AI risk management into four core functions: Govern (establishing organizational accountability and culture), Map (identifying context, stakeholders, and risk categories), Measure (assessing, analyzing, and tracking AI risks), and Manage (prioritizing and treating identified risks).

The RMF is deliberately non-prescriptive — it tells organizations what to think about, not what to do. This is intentional: different industries, use cases, and risk profiles require different controls. A one-size-fits-all prescription would be impossible to follow meaningfully across contexts ranging from hospital scheduling to content moderation.

Who Uses NIST?

The AI RMF has been widely adopted by Fortune 500 companies, federal agencies, and international organizations. It has also influenced the EU AI Act's structure and ISO international AI standards. Voluntary doesn't mean unimportant — the RMF shapes how US companies think about AI risk even without legal force.

NIST's Trustworthy AI Properties

The NIST AI RMF identifies seven properties of trustworthy AI: Accountable, Explainable, Interpretable, Privacy-Enhanced, Reliable, Safe, and Secure and Resilient. These properties form a vocabulary for AI risk assessment that has become influential beyond the RMF itself — appearing in corporate AI principles documents, government procurement requirements, and international standards discussions.

Lesson 1 Quiz

US AI governance and NIST
The NIST AI Risk Management Framework's four core functions are:
✓ Correct — Correct. Govern, Map, Measure, Manage — a vocabulary that has become widely adopted in AI risk management.
The NIST AI RMF organizes risk management into Govern, Map, Measure, and Manage functions.
The NIST AI RMF is deliberately non-prescriptive because:
✓ Correct — Correct. Non-prescriptive design allows the framework to work across contexts from hospital scheduling to content moderation.
The RMF is non-prescriptive by design — different contexts require different controls, and a one-size-fits-all prescription would be meaningless.
The US voluntary approach to AI governance reflects:
✓ Correct — Correct. The voluntary approach reflects a distinct governance theory — not the absence of governance, but a different theory about how to achieve it.
The US voluntary approach is a deliberate theory: that well-designed frameworks create more adaptive industry behavior than binding legislation.
Which of the following is among NIST's seven properties of trustworthy AI?
✓ Correct — Correct. Explainable is one of NIST's seven trustworthy AI properties, along with Accountable, Interpretable, Privacy-Enhanced, Reliable, Safe, and Secure/Resilient.
NIST's seven trustworthy AI properties are: Accountable, Explainable, Interpretable, Privacy-Enhanced, Reliable, Safe, and Secure and Resilient.

Lab 1 — NIST RMF Application

Apply the four-function framework to a real AI risk scenario

Your Task

Choose an AI system in a specific organizational context (hospital triage, bank fraud detection, school content filter, social media recommendation).

Walk through all four NIST AI RMF functions for that system: Govern, Map, Measure, Manage. For each, identify what the organization should specifically be doing — not what the framework says in general.

Name your AI system and organizational context. Start with the Govern function — what accountability structures and culture should this organization have in place?
AI Lab AssistantNIST RMF Analyst
Name your system and organizational context. Start with Govern. I will push you to be specific about what each function means for your particular case.
Module 3 · Lesson 2

Executive Orders and Federal Policy

How the US has tried to govern AI without comprehensive legislation — and the limits of that approach

In October 2023, President Biden signed Executive Order 14110 on AI. Major news coverage followed. AI safety advocates praised it. Companies began preparing compliance processes.

Less than two years later, President Trump revoked it. This is the fundamental instability of governing a transformative technology through executive orders alone.

AI Executive Orders: A Policy Tool with Limits

Without comprehensive legislation, US AI policy has been shaped significantly by Executive Orders — directives from the President that apply to federal agencies and, through procurement requirements, influence the private sector. Three EOs stand out:

EO 13960 (2020, Trump): Directed federal agencies to use AI in ways consistent with trustworthy AI principles — fairness, transparency, accountability. Created a framework for AI in government but with limited enforcement.

EO 14110 (2023, Biden): The most comprehensive US AI EO to date. Required safety testing and sharing of results for the most powerful AI models, directed NIST to develop safety standards, instructed agencies to develop guidance on AI use in their sectors, and created reporting requirements for AI-related national security risks.

EO 14179 (2025, Trump): Revoked EO 14110. Directed agencies to develop an AI Action Plan focused on AI leadership and competitiveness, with explicit skepticism toward regulations that could impede AI development.

The Problem with Executive Orders as AI Policy

Executive orders are inherently fragile policy instruments. They bind executive agencies but not Congress, courts, or private companies beyond procurement. They can be revoked by the next administration — as the 2025 revocation of EO 14110 demonstrated. They cannot create the legal certainty that investors and compliance professionals need. And they can create inconsistency: federal agencies may receive different AI governance directives from different administrations within a single decade.

Congressional Gridlock on AI

Multiple comprehensive AI bills have been introduced in Congress without passage. The political dynamics are complex: tech companies lobby intensively, different constituencies prioritize different risks, and legislators lack technical expertise to evaluate competing claims. The US has passed narrower AI provisions in defense and appropriations bills — but not the comprehensive framework the EU created.

Lesson 2 Quiz

Executive orders and federal AI policy
Which executive order revoked the Biden AI executive order?
✓ Correct — Correct. EO 14179 (2025, Trump) revoked EO 14110 (2023, Biden) and redirected AI policy toward competitiveness over safety requirements.
EO 14179 (2025) revoked EO 14110 (2023). EO 13960 was a 2020 Trump order on AI in government.
Executive orders are limited as AI policy tools because:
✓ Correct — Correct. The fundamental fragility of executive orders is that they can be revoked — creating policy instability across administrations.
Executive orders are inherently fragile — they bind executive agencies but not Congress or courts, and can be revoked by the next administration.
Biden's EO 14110 on AI required:
✓ Correct — Correct. EO 14110 required safety testing for powerful models, directed NIST to develop safety standards, and created sector-specific guidance requirements.
EO 14110 required safety testing of the most powerful models, NIST standard development, and agency-specific guidance — not universal mandatory certification or a new agency.
Congress has not passed comprehensive AI legislation primarily because:
✓ Correct — Correct. Congressional AI legislation faces a combination of structural obstacles — lobbying, competing priorities, expertise gaps, and partisan dynamics.
Congressional gridlock on AI reflects complex political dynamics — not constitutional limitations or the EU AI Act's reach.

Lab 2 — Executive Order Analysis

Evaluate the strengths and limits of EO-based AI governance

Your Task

Compare the governance goals of Biden's EO 14110 and Trump's EO 14179. What did each prioritize? What assumptions about AI risk and regulation underlie each approach?

Then assess: for the specific AI policy goal you care most about (safety, competitiveness, civil rights, innovation), which approach is more likely to achieve it — and what would make either approach more durable?

Start by describing what you see as the core difference between the two executive orders. I will probe your analysis.
AI Lab AssistantFederal AI Policy Analyst
Describe what you see as the core difference between the Biden and Trump AI executive orders. I will push you to examine the underlying assumptions and trade-offs.
Module 3 · Lesson 3

Sector-Specific Regulation

How the FDA, FTC, and financial regulators have extended their authority into AI

Before there was an AI Act, before executive orders, before most AI policy debates — there was the FDA clearing AI medical devices, the FTC pursuing unfair AI pricing, and the CFPB writing credit explainability guidance.

The US governance story isn't only about what Congress hasn't done. It's also about what existing regulators have quietly built.

The Sector-by-Sector Reality of US AI Governance

While Congress has not passed comprehensive AI legislation, existing sector-specific regulators have extended their authority into AI. This creates a patchwork of rules that differs dramatically by industry — sophisticated in some areas, largely absent in others.

FDA (Food & Drug Administration): Has regulated AI/ML-based software as medical devices for years. Approximately 700 AI-enabled medical devices have been cleared. The FDA requires clinical validation, post-market surveillance, and — increasingly — transparency about algorithm change management. It's the most mature US AI regulatory framework for any sector.

FTC (Federal Trade Commission): Has enforcement authority over unfair and deceptive trade practices, which extends to AI. Has taken action against companies for AI bias in pricing and credit, unlawful use of biometric data, and deceptive AI claims. The FTC's AI-related enforcement has accelerated since 2022, but it is reactive — responding to harms after they occur rather than requiring pre-deployment review.

CFPB (Consumer Financial Protection Bureau): Has issued guidance on AI in credit decisions, requiring that adverse action notices explain AI-driven credit denials in terms consumers can understand. Financial regulators more broadly have been aggressive on model risk management — requiring banks to document, validate, and monitor AI models used in credit, fraud, and risk decisions.

EEOC (Equal Employment Opportunity Commission): Has issued guidance on AI hiring tools, clarifying that employers using AI that discriminates based on protected characteristics face liability under existing civil rights law — regardless of whether the employer built the algorithm or purchased it.

The Gap Problem

Sector-specific regulation means significant domains have limited oversight. General-purpose consumer AI applications, political advertising AI, AI in entertainment and media, and AI in small business operations largely lack specific regulatory frameworks. These aren't necessarily low-risk domains — they simply lack sector regulators with clear AI jurisdiction.

Lesson 3 Quiz

Sector-specific AI regulation
The FDA's AI regulatory framework is significant because:
✓ Correct — Correct. The FDA's AI/ML medical device framework is the most developed sector-specific AI governance in the US.
The FDA's medical device framework, extended to AI, is the most mature US sector AI regulation — requiring clinical validation and post-market surveillance for cleared AI devices.
The FTC's approach to AI governance is primarily:
✓ Correct — Correct. The FTC enforces after harms occur — it does not require pre-deployment review of most AI systems.
The FTC's AI governance is reactive — it enforces against unfair and deceptive practices after harms occur, rather than requiring pre-deployment approval.
The EEOC's guidance on AI hiring tools clarifies that:
✓ Correct — Correct. EEOC guidance makes clear that purchasing rather than building a discriminatory AI tool does not shield employers from liability.
EEOC guidance clarifies that employers using AI that discriminates face liability under existing civil rights law — regardless of whether they built or purchased the algorithm.
The "gap problem" in US AI governance refers to:
✓ Correct — Correct. Consumer AI applications, political advertising AI, and many other domains lack specific oversight frameworks — not because they are low-risk, but because no sector regulator clearly owns them.
The gap problem means significant domains lack oversight because no existing sector regulator has clear AI jurisdiction — not a quality gap between US and EU frameworks.

Lab 3 — Regulatory Gap Mapping

Identify which AI systems fall through the cracks of US sector-specific regulation

Your Task

Choose an AI system operating in the United States that you believe may fall through the gaps of existing sector-specific regulation.

Make the case: which regulators have potential jurisdiction? What are the limits of that jurisdiction? What specific harms does the regulatory gap enable?

Name your AI system and your initial regulatory gap analysis. I will probe your jurisdictional claims.
AI Lab AssistantUS AI Regulatory Gap Analyst
Name your AI system and describe where you think it falls in the regulatory landscape. I will challenge your jurisdictional analysis.
Module 3 · Lesson 4

Voluntary vs. Mandatory

The policy debate at the heart of US AI governance — and what the evidence actually shows

Voluntary frameworks or binding rules. Industry-led or government-mandated. The debate recurs across every domain of technology governance, and AI is no different.

What is different is the scale of potential harm, the speed of deployment, and the growing body of evidence about what both approaches actually achieve in practice.

The Central Policy Question

The debate between voluntary and mandatory AI governance frameworks is not simply a regulatory philosophy dispute — it reflects genuinely different empirical predictions about how companies behave, how harms emerge, and how governance achieves its goals.

The Case for Voluntary Frameworks

Proponents argue that voluntary frameworks like the NIST AI RMF enable faster adoption — companies can move at their own pace rather than waiting for legal timelines. They allow for experimentation — companies try different approaches and share what works. They avoid regulatory capture — detailed prescriptive rules can be written by industry for industry. And they enable adaptation — as AI capabilities change, voluntary frameworks update more easily than enacted law.

The empirical question: do companies actually implement voluntary frameworks substantively, or do they treat compliance as a box-checking exercise with no meaningful risk reduction?

The Case for Mandatory Requirements

Proponents argue that voluntary frameworks create a race to the bottom — companies that invest heavily in AI governance face higher costs than competitors who implement frameworks superficially. Mandatory requirements create a level playing field. They also create accountability — voluntary commitments are unkept commitments. And they protect against collective action problems: a company might want to invest in safety but face investor pressure not to if competitors are not.

The empirical question: do mandatory frameworks meaningfully change company behavior, or do they generate documentation-heavy compliance exercises with no real risk reduction?

The Honest Answer

We do not yet have strong empirical evidence that either approach systematically achieves better AI safety outcomes. The EU AI Act is too new for meaningful outcome data. US voluntary frameworks have been insufficiently monitored to assess real compliance quality. Governance debates often rely more on theoretical predictions about behavior than observed outcomes.

State-Level Regulation

In the absence of federal AI legislation, states have moved. California has passed multiple AI transparency and accountability bills. Illinois requires notice and consent for AI in hiring. Colorado has passed AI insurance accountability legislation. Texas, Florida, and others have passed bills focused on government AI use. This creates a patchwork that companies with national operations must navigate — and that some argue provides the missing accountability layer while federal action stalls.

Lesson 4 Quiz

Voluntary vs. mandatory AI governance
A key argument for voluntary AI governance frameworks is:
✓ Correct — Correct. Proponents of voluntary frameworks emphasize adaptability, faster adoption, and avoidance of prescriptive rules that can be captured by industry.
The case for voluntary frameworks centers on adaptability and speed — companies can move faster and experiment without waiting for legal timelines.
A key argument for mandatory AI governance requirements is:
✓ Correct — Correct. The level playing field argument is central — voluntary frameworks may create competitive disadvantage for companies that comply substantively.
The mandatory case centers on creating a level playing field and accountability — voluntary commitments are often not kept when competitors don't keep them either.
The "race to the bottom" concern about voluntary frameworks refers to:
✓ Correct — Correct. The race to the bottom concern is that voluntary frameworks create incentives for superficial compliance rather than meaningful governance investment.
Race to the bottom means companies are incentivized to invest minimally in governance if competitors who comply superficially have lower costs — undermining the framework's purpose.
What has the evidence shown about voluntary vs. mandatory frameworks and AI safety outcomes?
✓ Correct — Correct. The EU AI Act is too new for outcome data; US voluntary frameworks have been insufficiently monitored. Governance debates rely more on theory than observed outcomes.
We lack strong empirical evidence that either approach systematically achieves better AI safety outcomes — the EU AI Act is new and US voluntary frameworks lack outcome monitoring.

Lab 4 — The Governance Debate

Argue both sides of the voluntary vs. mandatory AI regulation question

Your Task

Choose a specific AI domain: hiring AI, medical diagnosis AI, or AI content moderation on social platforms.

First, make the strongest case FOR voluntary governance frameworks in that domain. Then make the strongest case FOR mandatory requirements. Push for specifics — not general arguments.

Name your AI domain and start with the strongest case for voluntary governance. I will challenge you to make it concrete before moving to the mandatory side.
AI Lab AssistantAI Governance Policy Debate Coach
Name your domain and give me the strongest case for voluntary governance frameworks. I will push you to make it specific before we examine the mandatory side.

Module Test

15 questions · 80% to pass
The NIST AI RMF's four core functions are:
✓ Correct — Correct.
The NIST AI RMF functions are Govern, Map, Measure, and Manage.
The NIST AI RMF is non-prescriptive because:
✓ Correct — Correct.
Non-prescriptive design allows the framework to work meaningfully across very different contexts and risk profiles.
Executive Order 14179 (2025) primarily:
✓ Correct — Correct.
EO 14179 revoked Biden's AI executive order and shifted US AI policy priorities toward competitiveness and leadership over safety requirements.
Executive orders are limited as AI governance tools because:
✓ Correct — Correct.
Executive orders can be revoked by the next administration — as demonstrated in 2025. This makes them inherently unstable governance instruments.
The FDA's approach to AI governance is significant because:
✓ Correct — Correct.
FDA medical device regulation extended to AI is the most developed US sector AI framework, with real pre-deployment and post-market requirements.
The FTC's AI governance approach is best described as:
✓ Correct — Correct.
The FTC is reactive — it enforces against AI harms after they occur rather than requiring pre-deployment review.
EEOC guidance on AI hiring tools clarifies that:
✓ Correct — Correct.
EEOC guidance: employers are liable for discriminatory AI outcomes regardless of whether they built or purchased the tool.
The regulatory gap problem in US AI governance means:
✓ Correct — Correct.
The gap problem: many AI domains — consumer applications, political advertising, entertainment — lack specific oversight because no sector regulator clearly owns them.
Which of the following best represents the case for voluntary AI governance?
✓ Correct — Correct.
The voluntary case emphasizes speed, adaptability, and flexibility — companies can respond to changing AI capabilities without waiting for slow legislative processes.
The race-to-the-bottom concern about voluntary AI governance is that:
✓ Correct — Correct.
Race to the bottom: substantive governance investment creates competitive disadvantage when competitors implement frameworks superficially at lower cost.
What has evidence shown about voluntary versus mandatory AI governance outcomes?
✓ Correct — Correct.
We lack strong evidence that either approach systematically achieves better safety outcomes. The EU AI Act is too new; US voluntary frameworks lack outcome monitoring.
State-level AI regulation in the US has been:
✓ Correct — Correct.
Multiple states have moved on AI regulation — California, Illinois, Colorado, Texas and others have passed AI-related legislation while federal comprehensive law stalls.
Which sector has the most mature US AI regulatory framework?
✓ Correct — Correct.
The FDA's AI/ML medical device framework is the most developed US sector AI regulation, with approximately 700 cleared AI-enabled devices and established validation requirements.
CFPB guidance on AI in credit decisions requires:
✓ Correct — Correct.
CFPB guidance requires that adverse action notices explain AI-driven credit denials in understandable terms — not pre-approval or prohibition.
The NIST AI RMF's trustworthy AI properties include:
✓ Correct — Correct.
NIST identifies seven trustworthy AI properties: Accountable, Explainable, Interpretable, Privacy-Enhanced, Reliable, Safe, and Secure and Resilient.