L1
·
Quiz
·
Lab
L2
·
Quiz
·
Lab
L3
·
Quiz
·
Lab
L4
·
Quiz
·
Lab
Module Test
Module 7 · Lesson 1

No One Is in Charge — and That's the Problem

AI is the first technology powerful enough to reshape civilization, and the first with no global body actually responsible for its safety.
If something goes wrong with a global AI system, who do you call?

On November 30, 2022, a company called OpenAI released a chatbot named ChatGPT to the public. They expected maybe a million users in the first year. They got a million in five days. Within two months, ChatGPT had 100 million users — the fastest adoption of any technology in recorded history.

Governments around the world were caught flat-footed. The European Union had been working on AI rules since 2021, but its landmark law — the EU AI Act — would not be finalized until 2024. The United States had no federal AI law at all. China had some rules but only for narrow cases. No international body had any authority over the technology. The most transformative software ever released to the public had arrived, and the global response was: wait and see.

That gap — between a technology moving at the speed of software and governments moving at the speed of committees — is what this lesson is about.

Why Governing AI Is Harder Than Governing Anything Else

Think about how the world governs other dangerous things. Nuclear weapons are controlled through the Nuclear Non-Proliferation Treaty, signed in 1968. The materials needed to build a nuclear bomb — enriched uranium, plutonium — are physical, rare, and hard to hide. You can inspect a country's stockpile. You can count the warheads.

AI is different in almost every important way. The "materials" are data and code. Data is invisible. Code can be copied instantly and sent anywhere on Earth in seconds. A research lab in San Francisco can publish a paper, and within weeks, teams in a dozen countries have built their own version of the same system. You cannot put AI in a warehouse and count it.

This creates a fundamental problem for any government or international body trying to set rules: by the time the rule is written, the technology has already moved on. And if one country bans something, a lab in another country can simply keep going.

Governance
The systems of rules, laws, and agreements that decide who gets to do what, and what happens when things go wrong. A country's legal system is governance. So is a school's code of conduct. So — theoretically — is an international treaty.
Jurisdiction
The authority of a government or court over a particular place or subject. A city government has jurisdiction over the city. The problem with global AI: no one has jurisdiction over the whole planet.
What Exists Right Now

So what actually governs AI today? The honest answer is: a patchwork of incomplete tools, none of which were designed for this.

The EU AI Act (2024): The most ambitious attempt so far. The European Union passed a law that classifies AI systems by risk level — low, medium, high, and unacceptable. Systems deemed "unacceptable risk," like AI that manipulates people without their knowledge, are banned in EU countries. High-risk systems, like AI used in hiring or credit scoring, require audits. This is real law with real teeth — companies can be fined up to 7% of their global revenue for violations. But it only covers the EU's 27 countries and their 450 million people. The other 7.5 billion people on Earth are not covered.

Executive Order on AI (U.S., October 2023): President Biden signed a sweeping executive order instructing federal agencies to study AI risks, requiring companies developing the most powerful AI to share safety test results with the government, and directing agencies to draft new standards. But an executive order is not a law — it can be reversed by the next president. And in fact, President Trump rescinded it on his first day in office in January 2025.

The Bletchley Declaration (November 2023): Twenty-eight countries — including the US, UK, China, France, Germany, and others — met at Bletchley Park in England (the famous World War II codebreaking site) and signed a statement agreeing that "frontier AI" poses serious risks and that international cooperation is needed. But a declaration is not a treaty. It has no enforcement mechanism. It is, essentially, an agreement to agree someday.

The Enforcement Gap

Here is the core tension in global AI governance: the countries with the most to gain from AI — the US, China, UK — are also the ones most resistant to international rules that might slow them down. And without those countries fully on board, any global framework is mostly symbolic.

The Precedent Problem: What We've Tried Before

When humans have faced global technological risks before, they sometimes managed to cooperate. The Montreal Protocol (1987) phased out chemicals that were destroying the ozone layer — and it worked. The Chemical Weapons Convention (1993) banned an entire category of weapons, and 193 countries have signed it. So we know international cooperation on dangerous technology is possible.

But there are important differences. Ozone-destroying chemicals were made by a small number of companies, which made them easier to regulate. Chemical weapons are almost universally hated — there's no powerful lobby saying "we need more nerve agents for the economy." AI is different: it has enormous commercial value, and every major power sees it as central to economic competitiveness and military strength. The incentives to race ahead are enormous. The incentives to cooperate are real but weaker.

This is the genuine tension you need to sit with: the tools that worked before may not work here, and the tools we'd need for this situation don't fully exist yet.

You Now See What Most People Miss

When you hear someone say "they should just regulate AI," you now understand the actual difficulty. Who is "they"? Regulate what, exactly — the models, the training data, the applications? And how do you enforce rules on something invisible that crosses borders in milliseconds? Most people who say "just regulate it" have never thought about these questions. You have.

An Ethical Question Without a Clean Answer

Here is a real tension that experts genuinely disagree about:

Strict global AI governance would probably slow AI development. That might prevent some catastrophic risks. But it would also slow the development of AI that could cure diseases, help farmers grow food in a changing climate, or educate children in places that don't have enough teachers. Every year of delay has a cost in lives that might have been saved or improved.

So: how much risk is acceptable in exchange for how much speed? And who gets to decide — the people building the AI, the governments of powerful countries, or the people who will be most affected by it (who are often in countries with the least political influence)?

There is no clean answer. That discomfort is not a failure — it is what honest thinking about hard problems feels like.

Lesson 1 Quiz

Four questions. Test your reasoning, not your memory.
Why is governing AI harder than governing nuclear weapons, even though both are potentially dangerous?
Exactly. Nuclear materials are physical, rare, and verifiable. AI is code — weightless, copyable, borderless. That physical difference makes all the governance tools we developed for weapons much harder to apply.
Think about what makes nuclear governance possible in the first place. It's not just the severity of the danger — it's that you can physically count and inspect the materials. What's different about AI?
The Bletchley Declaration (2023) was signed by 28 countries. Why is it still considered a weak form of governance?
Right. A declaration is basically a press release with signatures. It signals intention but creates no legal obligation. Without an enforcement mechanism, "agreeing that something is serious" doesn't actually change what anyone does.
Check the lesson again. The issue isn't who signed it or what it said — it's the legal status of a declaration versus a binding treaty. What's the difference between those two things?
A new country develops a powerful AI system that other nations find dangerous. The EU AI Act bans that type of system. What is the most likely limitation of this situation?
Correct — and this is the jurisdiction problem in a nutshell. The EU can regulate what happens inside the EU. It cannot tell a lab in a non-EU country what to build. This is why regional laws, no matter how well designed, can't solve a global problem alone.
Think about what jurisdiction means. The EU AI Act is law — but whose law? In what territory does it apply? Does the EU have authority over what happens in another country?
The lesson presents a genuine ethical tension about the speed of AI governance. Which of the following best describes that tension?
Yes. This is what makes it a genuine ethical question rather than just a political one. Both sides of the argument involve real human welfare. Faster AI: more potential cures, tools, improvements. Slower regulation: more risk of harm. The honest answer is that reasonable people disagree.
The lesson is pointing at something more specific — a trade-off between two things that both have value. What does slowing AI governance cost? What does speeding it up risk?

Lab 1: The Treaty Designer

You're drafting the framework for a global AI safety agreement. Your peer advisor will challenge every choice you make.

Your Role: Policy Architect

You've been asked to sketch the core structure of a global AI safety treaty. Your advisor — another policy architect who's seen previous international agreements fail — will push back on your ideas, ask hard questions, and refuse to let you off the hook with vague answers.

This isn't about getting the "right" answer. It's about defending a position and thinking through consequences.

Start by telling your advisor: What is the single most important thing a global AI safety treaty must include, and why? Then be ready to defend it.
Policy Advisor
Lab 1
Alright. You've been given thirty minutes to pitch the core of a global AI safety treaty to a room of diplomats. I've seen three of these fail — the Bletchley Declaration, a proposed UN AI body in 2024, and a bilateral US-China framework that never got past the first draft. I'm not impressed by good intentions. Tell me: what's the one non-negotiable element your treaty needs, and why should any major AI power actually sign it?
Module 7 · Lesson 2

The Race Dynamic: Why Countries Compete Instead of Cooperate

When two powerful players both believe winning is better than sharing, even a dangerous race can feel like the rational choice.
If everyone agrees AI could be dangerous, why does no one slow down?

In March 2023, the U.S. government added more Chinese AI companies to its trade restriction list — meaning American companies were forbidden from selling them advanced computer chips. The specific target was Huawei's new chips and a company called Biren Technology. The reasoning: advanced chips are what make powerful AI possible, and the U.S. wanted to ensure China could not close the gap in AI capability.

China's response was to accelerate its own chip manufacturing. Within months, Huawei released a new phone with a chip that analysts said shouldn't have been possible given the restrictions. By late 2023, China was openly discussing plans to become the world's leading AI power by 2030.

Meanwhile, in the U.S., Congress was holding hearings about AI safety — but the dominant message from industry and many officials was: we can't slow down or China will win. Both sides were accelerating. Both sides cited the other's acceleration as the reason they had to accelerate. And no one was waiting for a safety review.

Understanding the Race: Game Theory in One Paragraph

There's a concept in mathematics and economics called game theory — the study of how rational actors make decisions when what they should do depends on what others do. One of its most famous ideas is the "prisoner's dilemma."

Here's the short version: Imagine two people who would both be better off if they cooperated. But if one cooperates and the other doesn't, the one who defected (didn't cooperate) does even better, and the cooperator is left worse off. So even if both players know cooperation is the best mutual outcome, each one is tempted to defect — because they're afraid the other will. The result: both defect, both end up worse than if they'd cooperated, and neither is willing to go first.

The US-China AI race fits this pattern almost perfectly. Both countries would arguably be better off in a world where AI development is slower and safer. But if the US slows down and China doesn't, the US fears falling behind in a technology it sees as strategically essential. And China fears the same thing in reverse. So both accelerate, even though neither side is thrilled about the risks.

Prisoner's Dilemma
A situation where two parties would both benefit from cooperating, but each party has an individual incentive to defect (not cooperate), making it hard to achieve the mutually beneficial outcome. Named after a thought experiment about two prisoners deciding whether to betray each other.
It's Not Just Two Countries

The US-China dynamic gets the most attention, but the race is broader. The United Kingdom, France, Canada, South Korea, India, Israel, and the UAE are all investing heavily in domestic AI development. Private companies — OpenAI, Google DeepMind, Anthropic, Meta, Mistral, Baidu — operate with their own competitive pressures, often moving faster than any government. A company that releases a powerful model first gains users, revenue, talent, and influence. Every month of delay is a competitive cost.

This means the race is not just between nations — it's between companies, between labs, between research teams inside the same organization. In 2023, Google reportedly rushed to release its Bard chatbot because ChatGPT was taking market share. Some Google engineers later said the product was not ready. The competitive pressure overrode the cautious timeline.

What you're seeing is a structure — a set of incentives built into how competition works — that pushes toward speed regardless of what any individual actor wants. Even people who genuinely believe AI is dangerous find themselves inside an institution that has to keep up or fall behind.

Why This Matters for Safety

Safety testing takes time. Red-teaming — having experts try to find dangerous flaws in an AI system — takes months. Interpretability research (trying to understand why an AI makes the decisions it does) is slow, difficult work. In a race, all of these feel like delays. And when everyone is racing, safety work gets compressed, deprioritized, or done in parallel with deployment rather than before it.

Has Anyone Tried to Slow the Race?

In March 2023, an open letter signed by over 1,000 people — including prominent AI researchers, Elon Musk, and Apple co-founder Steve Wozniak — called for a six-month pause on training AI systems more powerful than GPT-4. The letter cited risks to society and said the pause was needed to develop safety protocols.

The pause did not happen. OpenAI, Google, Anthropic, and Meta all kept working. The letter's signatories disagreed significantly with each other about why a pause was needed and what should happen next. Within weeks, the conversation had moved on.

The pause letter episode illustrates something important: voluntary restraint is very hard to achieve in a competitive environment. Even people who agreed something was dangerous couldn't coordinate a slowdown, because each lab feared that only they would stop while others kept going.

This is exactly why external governance — rules that apply to everyone equally — is theoretically valuable. If all labs are required to pause, no one loses a competitive advantage by pausing. But getting to that rule requires international agreement, which brings us back to the problems in Lesson 1.

The Structural Insight

You can now see something that gets lost in headlines: the AI race isn't primarily about greed or recklessness. It's about structure. The incentives built into competition — between companies, between nations — push toward speed even when individuals inside those organizations would prefer to go slower. Understanding this changes how you evaluate calls for "just be more responsible." Responsible behavior is hard when the structure punishes it.

An Ethical Question to Sit With

If voluntary restraint doesn't work, and binding international agreements are politically very hard to achieve, what is left? Some researchers argue for unilateral slowdowns — one country or company deciding to go slower regardless of what others do, as a moral statement and a demonstration that it's possible. Others argue this is naive: you just fall behind while others take the risks anyway, and the world doesn't become safer.

Who should bear the cost of slowing down, if anyone should? The companies that would lose market share? The countries that would fall behind? Or is the cost of not slowing down — potential catastrophic risk — larger than any of those costs? There is no consensus on this. The people arguing about it are serious, informed, and genuinely uncertain.

Lesson 2 Quiz

Think about structure, incentives, and what the race dynamic actually means.
In 2023, the U.S. restricted chip exports to China to slow its AI development. China responded by accelerating its own chip manufacturing. This sequence best illustrates which concept?
Exactly. The restriction was meant to slow China down — instead it motivated China to develop its own capabilities faster. The action that was supposed to create safety created more urgency to compete. That's the race dynamic in action.
Look at the sequence: US restricts → China accelerates. Did the restriction cause cooperation or escalation? What does that tell you about the effectiveness of unilateral moves in a competitive environment?
The 2023 open letter calling for a six-month AI pause failed to actually pause development. What does this most clearly reveal about the limits of voluntary restraint?
Right. The letter wasn't ignored because people disagreed with its premise — it was ignored because each lab feared being the only one to pause. This is exactly the structure of the prisoner's dilemma: the individually rational move undermines the collectively good outcome.
Think about what stopped each individual company from complying. Was it disagreement with the idea, or fear of what competitors would do if they paused? What does that fear reveal about the limits of voluntary approaches?
You work at an AI company that genuinely cares about safety. Your team wants six more months to test your new model. Your CEO says: "If we wait, our competitor releases first and we lose the market." This scenario illustrates that reckless AI development is primarily caused by:
Exactly — and this is the key structural insight from the lesson. The people inside these decisions often know the risks. The competitive structure still pushes toward speed. That's why "just be more responsible" misses the point. The problem isn't only attitude — it's incentives.
In this scenario, does your CEO not care about safety? Probably not — the issue is the competitive situation they're in. What created that pressure? Is it about individual character, or about the structure of competition?
Why would binding international rules (that apply to all labs equally) theoretically solve the coordination problem that voluntary restraint cannot?
Correct. The value of binding rules in a competitive environment is that they level the playing field. When everyone has to follow the same constraint, no one is punished for being cautious. That's the theoretical advantage. The practical challenge is getting those rules agreed to and enforced — which is where governance comes back in.
Think about what voluntary restraint fails to eliminate: the fear that you stop but others don't. How does a rule that applies to everyone — with enforcement — change that fear? Does it eliminate it?

Lab 2: The Race Analyst

You're analyzing a competitive AI race scenario. Your peer will pressure-test every claim you make about what's driving it.

Your Role: Strategic Analyst

Two hypothetical countries — Aldoria and Ventis — are both developing powerful AI systems. Aldoria announces a voluntary six-month safety review. Ventis keeps going. You need to analyze what happens next and what, if anything, could have been done differently.

Your advisor will challenge whether your analysis correctly identifies causes versus symptoms, and whether your proposed solutions actually address the structural problem.

Begin: Aldoria paused, Ventis didn't. What is Aldoria's next move, and what does this situation tell us about the limits of voluntary safety measures?
Strategic Advisor
Lab 2
Aldoria just announced its pause. Ventis released a press statement two hours later saying they're accelerating. I've seen this exact sequence play out with arms control agreements, environmental treaties, and trade deals. Before you tell me what Aldoria should do — I need you to be precise. Is Aldoria's problem a lack of willpower, a lack of information, or a structural problem with the incentives? Because the answer changes everything about what "do next" means.
Module 7 · Lesson 3

The Institutions: What Actually Exists and What It Can Do

Governance isn't just laws — it's institutions, norms, standards, and the slow build of trust between parties who don't fully agree.
What does a functioning global AI safety institution actually look like?

In October 2024, the UK government opened the doors of the world's first national AI Safety Institute — a government body whose entire job is to evaluate frontier AI models for dangerous capabilities before they are released to the public. The Institute, led by Yoshua Bengio (one of the three researchers credited with creating modern deep learning) on its scientific advisory board, started doing something no government had done before: actually testing AI systems for dangerous behaviors like the ability to help create bioweapons or assist cyberattacks.

The US launched its own AI Safety Institute shortly afterward, housed within the National Institute of Standards and Technology (NIST). Within months, the UK and US AI Safety Institutes had signed a formal agreement to share testing methodologies and coordinate evaluations. South Korea and Canada announced they were building their own institutes. A model was emerging — not a single global body, but a network of national institutions learning to work together.

It was the most concrete governance infrastructure the world had produced. It was also, critics noted, entirely voluntary. Companies shared models with these institutes when they chose to. And then in 2025, the US Institute was significantly cut during the Trump administration's federal restructuring. The fragility of the system became visible almost immediately.

The Landscape of Institutions

When people say "AI governance," they often imagine a single powerful body — a kind of "UN for AI" — making binding rules that everyone follows. That doesn't exist. What exists instead is a complex, overlapping set of organizations with different powers, different mandates, and different relationships with each other. Understanding this landscape is what lets you evaluate any governance proposal clearly.

National AI Safety Institutes: The UK and US models above. Government-funded bodies that evaluate AI models, publish safety standards, and advise policymakers. Strength: they have real technical capacity and can do actual evaluations. Weakness: they're funded by governments that can cut them, and participation by AI companies is currently voluntary.

The United Nations: The UN Secretary-General's office released a report in 2023 calling for a new international AI body. The UN's existing agencies — like the International Telecommunication Union (ITU) — have started working on AI standards. But the UN works by consensus, meaning any one of its 193 member states can block action. Historically, this makes binding agreements very slow. The UN is better at building norms (shared expectations about behavior) than hard rules.

The G7 and G20: Groups of the world's major economies that meet regularly and make joint statements. The G7 AI Principles (Hiroshima AI Process, 2023) set out a framework for responsible AI that member countries agreed to promote. But again: principles are not laws. They don't bind anyone. They're expressions of shared values that may — or may not — get translated into actual policy.

Technical Standards Bodies: Organizations like the International Organization for Standardization (ISO) and the National Institute of Standards and Technology (NIST) develop technical standards — precise specifications for how things should be built or tested. These are less politically exciting but often more practically powerful. When ISO publishes a standard for AI risk management (which it did in 2023), companies that want to sell in international markets have a strong incentive to comply even without a law requiring it.

Norm
A shared expectation about how actors should behave. Norms don't have legal force, but they matter: countries and companies that violate widely held norms face reputational costs, diplomatic pressure, and sometimes sanctions. Most international behavior is governed more by norms than by hard law.
Technical Standard
A precise specification that defines how something should be measured, tested, or built. If a standard requires AI systems to pass certain safety evaluations before deployment, companies often follow it even without a law — because customers, insurers, and business partners expect compliance.
The IAEA Model: Can It Work for AI?

The most frequently proposed model for global AI governance is the International Atomic Energy Agency — the IAEA. Founded in 1957, the IAEA is a UN body that simultaneously promotes peaceful uses of nuclear energy and monitors countries for nuclear weapons development. Its inspectors physically visit nuclear sites. Its reports can trigger UN Security Council action. It has real teeth.

Several serious proposals have called for an "IAEA for AI." The idea: an international body with the authority to inspect AI development facilities, require transparency about the most powerful models, and flag dangerous systems. The Biden administration supported the idea in 2023. Several European governments expressed interest.

The counterarguments are also serious. Nuclear materials are physical and verifiable — inspectors can measure uranium enrichment. AI training runs are computational and invisible. A lab could run a secret training job in the cloud with no physical footprint to inspect. The verification problem that makes nuclear governance possible is much harder for AI. Additionally, the IAEA's authority comes from the fact that building a nuclear bomb is fantastically expensive and requires rare materials. AI requires only electricity and data — which are everywhere.

This doesn't mean the IAEA model is useless as a reference. It means the AI version would need to rely more on software monitoring, self-reporting, and whistleblower protections — and less on physical inspection. Whether that's sufficient is an open question.

Institutional Reality Check

Every functioning international institution was considered impossible before it existed. The IAEA, the World Trade Organization, the International Criminal Court — all faced predictions that powerful nations would never accept that kind of constraint on their sovereignty. They were partly right: the US still hasn't ratified the ICC's founding treaty. But partial institutions, imperfect institutions, still shape behavior. The question isn't "perfect governance or nothing" — it's "what can we build that's better than what we have?"

What Institutions Actually Do Well

Here's something that gets underappreciated in debates about AI governance: institutions don't need perfect enforcement to matter. What they do well is build shared definitions, create accountability norms, and give responsible actors cover to do the right thing.

When a safety standard exists, a company that wants to move carefully can point to it and say: "We're complying with international standards." That gives internal safety advocates something to cite. It gives investors something to evaluate. It gives journalists something to compare companies against. This soft accountability can shift behavior even without legal penalties.

The UK AI Safety Institute's evaluations work this way. No law requires companies to submit models for testing. But when OpenAI, Google DeepMind, and Anthropic all agreed to do so, it created an expectation. A company that refused would stand out and face questions about what it was hiding. The norm did work that a law hadn't yet been written to do.

You're now in a position to evaluate AI governance proposals with real nuance. When someone announces a new AI safety body, you can ask: Does it have enforcement power or just advisory power? Is participation mandatory or voluntary? Who funds it, and can that funding be cut? Who does it have authority over — just domestic actors or international ones too? These questions cut through the press releases.

Knowing This Changes How You Read the News

Every headline about a new AI safety initiative, framework, or agreement is now legible to you in a way it wasn't before. You can classify it: Is this a law with enforcement, a voluntary standard, a declaration of principles, or a research body? That classification tells you how much it will actually change behavior. Most people reading those headlines can't make that distinction. You can.

An Ethical Question

The institutions described in this lesson were largely designed by the world's most powerful countries — the US, UK, EU, and their allies. The standards they produce reflect their priorities, their threat models, and their values. Countries that weren't in the room when these standards were written — most of Africa, much of Latin America, South and Southeast Asia — now have to decide whether to adopt frameworks they had no role in shaping.

Is that the right way to build global AI governance? If the alternative is no governance at all — because getting everyone in the room first would take decades — maybe imperfect, powerful-country-led governance is better than nothing. Or maybe governance designed without the global majority will fail to address their actual concerns and will never be seen as legitimate. Both of these are defensible positions. The people making these decisions are genuinely uncertain.

Lesson 3 Quiz

Apply the institutional landscape to new scenarios.
The UK AI Safety Institute began testing AI models before they were publicly released. Companies participated voluntarily. What type of governance mechanism does this represent, and what is its key vulnerability?
Correct. The AISI created a norm: major labs participate in pre-deployment evaluations. That norm has real effects — but it depends on participation staying voluntary and funding staying stable. The lesson showed both vulnerabilities clearly: the Institute was cut, and participation was never legally required.
Think about what kind of mechanism requires voluntary participation but still creates pressure to comply. Is it a law? A norm? A technical standard? What's the specific mechanism — and what happens if a company or government simply decides to stop?
A proposal calls for creating an "IAEA for AI" — an international body with inspectors who can visit AI labs. Critics argue the analogy breaks down. What is the strongest argument against this analogy?
Exactly. The IAEA model works because you can count centrifuges and measure uranium. AI training can happen in a data center with no obvious physical signature. The verification tools that make nuclear governance possible don't transfer cleanly to AI — which means an AI version would need fundamentally different enforcement mechanisms.
The IAEA's power comes from a specific capability: physical inspection and material measurement. What is the equivalent for AI? Is there one? That's where the analogy starts to strain.
A new ISO standard for AI risk management is published. No law requires companies to follow it. Why might companies still comply?
Right. This is the "soft accountability" mechanism described in the lesson. Standards create a reference point. Enterprise customers, investors, and insurers use that reference point to evaluate companies. Non-compliance isn't illegal — but it becomes something you have to explain, which creates real pressure.
ISO doesn't have the power to fine companies or make laws. So why would a company follow a standard? Think about who else cares about whether a company meets a standard — customers, investors, partners — and what those parties might do with that information.
The lesson raises the concern that major AI governance frameworks were designed primarily by the US, UK, and EU. Imagine you're advising a government in Southeast Asia considering whether to adopt these frameworks. What is the most honest challenge to raise?
This is the real tension the lesson was pointing at. There's a genuine trade-off between the legitimacy of inclusive governance and the practicality of having any governance quickly. That trade-off doesn't resolve cleanly — which is why it's an ethical question worth sitting with, not a problem with a tidy solution.
Think about what "your country wasn't in the room" actually means for governance quality and for the willingness of your population to trust and follow those rules. Then think about what refusing all outside frameworks would mean for safety. Is either option clean?

Lab 3: The Institution Auditor

Evaluate a proposed AI governance body using the framework you now understand. Your advisor will demand specifics.

Your Role: Independent Evaluator

A new international body has been announced: the "Global AI Accountability Council" (GAAC). It will issue safety guidelines, accept voluntary model submissions from AI companies, and publish annual reports. It is funded by membership fees from participating countries. The US, EU, and China are all listed as founding members. No enforcement powers are mentioned.

Your advisor wants your assessment: how strong is this institution, and what specific gaps concern you most?

Start by classifying this institution. Is it a binding treaty body, a voluntary standards organization, a norm-setting forum, or something else? Then identify the single biggest structural weakness in the design.
Governance Advisor
Lab 3
I've read the GAAC announcement. I've seen about thirty of these in the past three years, and most of them are press releases with a logo. Before I get excited, I need your classification — is this a binding treaty body, a norm-setting forum, a voluntary standards org, or a funding mechanism? Use what you know to place it precisely. And then I want the single structural weakness that, if not addressed, makes this thing useless in a real crisis.
Module 7 · Lesson 4

What Comes Next — and Where You Fit In

The people who will shape AI governance in 2035 are teenagers right now. That includes you.
If the current system isn't enough, what would actually be better — and who gets to build it?

In May 2023, the UN Secretary-General António Guterres stood before an audience of diplomats and made a striking request. He proposed creating a new international AI body modeled on the International Atomic Energy Agency and the Intergovernmental Panel on Climate Change — a body that would both assess risks and set international norms. He noted that the people most affected by AI's future were not in the room: they were young people in countries that had no seat at the table.

The proposal went into committee. It has not emerged as a functioning institution. But it opened a question that has not closed: if not a UN body, then what? Every alternative — national institutes, industry coalitions, technical standards bodies — has gaps. The governance architecture of AI in 2035 has not been written yet. The debates happening right now will determine its shape. The people who are learning this material today are precisely the people who will be in those debates — as voters, as engineers, as lawyers, as policymakers, as journalists, or as citizens who hold all of those people accountable.

Three Possible Futures

Based on what serious researchers and policymakers are actually discussing, here are three plausible directions for global AI governance — each with real trade-offs.

Scenario A: The Network Model. No single global body. Instead, national AI safety institutes cooperate — sharing testing data, coordinating standards, and running joint evaluations. The US, UK, EU, Japan, South Korea, and others form a kind of club with shared practices. China and others may have parallel institutions that occasionally exchange information. This is closest to where things are heading now. Its strength: it's buildable. Its weakness: it leaves out most of the world and has no mechanism for situations where the major powers disagree.

Scenario B: The UN Agency Model. A new international body — perhaps growing out of existing UN structures — is given a formal mandate to evaluate frontier AI, publish public assessments, and flag systems that pose international risks. Countries retain sovereignty but agree to the evaluation process. This would require a treaty, which requires Senate ratification in the US (historically difficult), buy-in from China, and a functioning consensus among deeply competitive nations. Its strength: legitimacy. Its weakness: political feasibility and the verification problem.

Scenario C: The Market-Led Model. Governance happens primarily through market forces. Insurance companies require AI systems to pass safety evaluations before issuing liability coverage. Enterprise customers demand compliance with voluntary standards as a condition of purchase. Investors price safety performance into valuations. Governments focus on liability law — if an AI system causes harm, the deployer is legally responsible. This model requires no international treaty. Its weakness: market forces tend to reward profit, not safety, and don't protect the people with the least market power.

Sovereignty
A country's authority to govern itself without outside interference. International governance always involves tension with sovereignty — countries agreeing to be bound by outside rules gives up some of that authority. How much sovereignty to trade for how much safety is a core political question.
The Inclusion Problem

In 2024, a survey of AI governance forums found that the Global South — countries in Africa, Latin America, South and Southeast Asia — accounted for roughly 12% of participants in major AI governance discussions, despite representing about 70% of the world's population. This is not an accident. It reflects which countries have the most AI companies, the most researchers, and the most diplomatic resources to engage in specialized international forums.

But who has the most to gain or lose from AI is a different question. Many of the applications being developed — AI for crop disease detection, for early disease surveillance, for language translation in underserved languages — are most consequential for populations in lower-income countries. And many of the harms — algorithmic discrimination in credit and hiring, AI-powered surveillance sold to authoritarian governments, automated systems replacing low-wage jobs with no social safety net — land hardest on populations with the least political power.

The institutions being built right now will set the norms for decades. If those institutions are designed without input from most of the world's population, they will reflect a subset of values and concerns. Whether that's better than no institutions is genuinely debatable. What's not debatable is that the design process is happening now, and who's in the room matters.

What "Being in the Room" Actually Means

International governance involves many types of participation: heads of state at summits, career diplomats in UN committees, technical experts drafting standards, civil society organizations commenting on proposed rules, journalists whose reporting shapes public pressure, lawyers who interpret and enforce agreements, and researchers whose work gives policymakers the information they need to act. You don't have to be a diplomat to influence governance. You have to understand what's happening well enough to contribute something useful to those processes — and to hold the people in them accountable.

The Specific Work That Remains Undone

Here, concretely, is what serious people in this field say needs to be built and hasn't been yet:

A shared definition of "frontier AI": Governance frameworks keep referring to the most powerful AI systems as "frontier AI" or "advanced AI" — but there is no agreed international definition of which systems qualify. Without a shared definition, there's no way to know what the rules actually apply to.

Verification mechanisms: A way to confirm that AI systems have actually been tested against claimed safety standards, and that reported training compute is accurate. This requires technical tools that don't fully exist yet — software auditing methods, hardware tracking systems, and agreed measurement protocols.

Liability frameworks: Clear legal rules about who is responsible when an AI system causes harm. If an AI medical diagnostic system gives wrong advice and a patient is harmed, who is liable — the hospital that deployed it, the company that built it, or the company that trained the underlying model? Most legal systems have not caught up to this question.

Inclusive standard-setting: Mechanisms for countries that lack major AI companies or large research budgets to meaningfully participate in shaping the standards that will affect their populations.

These are not unsolvable problems. They are work. Hard, slow, mostly unglamorous work that will be done by people who understand both the technology and the political and legal systems around it. Some of those people have not started their careers yet.

Where You Are in This Story

You now understand the full landscape: why global AI governance is structurally harder than previous international challenges, what institutions exist and what they can and can't do, what the race dynamic is and why it persists, and what the specific gaps are in current frameworks. That understanding is not common. Most adults making decisions about AI — in government, in business, in journalism — don't have all of this clearly in their heads. You do. That's not a small thing. The question is what you build with it.

A Final Question Without an Answer

The hardest governance question in this field is not technical. It's this: AI governance requires powerful countries to accept constraints on something they see as central to their economic and military power. History suggests that happens very slowly, and usually only after a crisis.

But with some categories of AI risk — systems that could enable the creation of biological weapons, or that could operate faster than any human institution can respond — waiting for a crisis may mean waiting until after the harm has already been done.

So: is it possible to build adequate governance before a catastrophe, or does history tell us we will need the catastrophe first? And if you believe we can do better than history suggests, what would need to be different this time?

That question is live. It doesn't have a consensus answer. The debate is happening now, and the people contributing to it include people your age who got serious about understanding this early.

Lesson 4 Quiz

Evaluate the scenarios and apply the full module's reasoning.
The "Network Model" of AI governance — where national safety institutes cooperate and share standards — is closest to what's actually developing now. What is its most significant structural weakness?
Correct. The network model is pragmatic and buildable — but it's a club of powerful countries. When those countries agree, it works reasonably well. When they don't — or when a country outside the club is involved — the model has no resolution mechanism. And the populations most affected by AI may not be in the club.
Think about who's in the network and who isn't. What happens when two network members disagree? What happens when an actor outside the network is the problem? The weakness isn't about legality or slowness — it's about coverage and conflict resolution.
The "Market-Led Model" uses insurance requirements, customer demand, and investor pressure to drive AI safety compliance. A critic argues this model systematically fails one group. Who is that group, and why?
Right. Markets are good at delivering what paying customers want. They are structurally poor at protecting people who aren't paying customers — or whose harms don't register as legal liability. The people most vulnerable to AI harms are often the people with the least ability to exert market pressure.
Think about how market pressure actually works. Who can withhold a purchase? Who can demand better terms from an insurer? Who does the market actually listen to? Then think about who doesn't have that power — and whose harms that model may never address.
The lesson lists specific governance gaps that remain unsolved, including the lack of a shared definition of "frontier AI." Why is this definitional problem a governance problem, not just a technical one?
Exactly. Any law or treaty that says "these rules apply to frontier AI systems" immediately runs into the question: which systems qualify? Without a precise, agreed definition, every actor has an incentive to argue their system is just below the threshold. The definitional problem is the jurisdictional problem in miniature.
Think about how laws work. A law that says "you must follow these rules if you build a frontier AI system" requires a definition of "frontier AI" to be enforceable. Without that definition, what happens when a company disagrees about whether its system qualifies?
The lesson ends with the question: Can we build adequate governance before a crisis, or do we need the crisis first? Based on everything in this module, which argument is most defensible?
This is the honest answer. Historical precedent is mixed — sometimes crises produce governance breakthroughs, sometimes they produce bad governance, sometimes the crisis is too severe for recovery. For some AI risk categories, there may be no recovery from the first serious incident. That asymmetry is what makes pre-crisis governance important and hard. You're not supposed to be certain here. Uncertainty is the accurate position.
Look at what history actually tells us: some governance happened before crises (the Montreal Protocol, partly), some required a crisis (nuclear governance accelerated after Hiroshima). And for some risks, "wait for the crisis" isn't viable because the crisis is unrecoverable. Which of those patterns applies to different AI risk categories?

Lab 4: The Governance Architect

Design a governance proposal for 2035. Your advisor will challenge every assumption about what's politically possible.

Your Role: Senior Policy Architect

It's 2027. A major international AI incident has occurred — a powerful AI system was used to generate convincing disinformation that destabilized a national election. No existing framework had authority to respond. You've been asked to propose a governance architecture for the next ten years that addresses this specific failure while being politically achievable.

Your advisor has been through the UN system and seen three previous governance frameworks collapse. She will push you on feasibility, not just idealism.

Start by identifying which of the three governance models from Lesson 4 — Network, UN Agency, or Market-Led — your proposal is closest to. Then explain the single most important thing your framework must do differently from what exists now, and why you believe it can survive political resistance from major AI powers.
Senior Policy Advisor
Lab 4
I was in the room for the 2024 Bletchley follow-up meetings and watched three good frameworks die because no one had answered the sovereignty question. You're not going to make that mistake. Before you give me the full proposal, tell me: which of the three models are you starting from, what's the one thing it must do that current frameworks don't, and — most importantly — why would the US and China both sign it? Give me the honest version, not the optimistic one.

Module 7 Test

15 questions across all four lessons. Score 80% or higher to pass.
1. ChatGPT reached 100 million users faster than any technology in history. What governance challenge did this speed primarily expose?
Correct. The EU AI Act was already in progress but not finalized. The US had no federal law. The deployment-governance gap was real and immediate.
Think about the timeline: ChatGPT deployed November 2022. EU AI Act finalized 2024. No US federal law. The gap between deployment and governance is the core issue.
2. What is the primary reason nuclear weapons are easier to govern internationally than AI systems?
Right. The physical verifiability of nuclear materials is what makes IAEA-style inspection possible. That tool doesn't transfer cleanly to AI.
The key is verifiability. How do nuclear inspectors confirm compliance? Can that same method work for AI?
3. The Bletchley Declaration was signed by 28 countries in 2023. Why does this represent a weak form of governance?
Correct. A declaration is a statement of intent, not a legal obligation. Without enforcement, signatories can ignore it without consequence.
What's the legal status of a declaration versus a treaty? What happens if a signatory simply doesn't follow through?
4. The prisoner's dilemma concept from Lesson 2 best explains why:
Exactly. The dilemma is structural: the individually rational move (keep racing) produces a collectively worse outcome (everyone races into risk). Even actors who prefer caution find themselves inside systems that punish it.
The prisoner's dilemma is about rational individual choices producing bad collective outcomes. How does that apply to the decision each country or company makes about pace?
5. In 2023, over 1,000 people — including prominent researchers — signed an open letter calling for a six-month pause on AI development. Development did not pause. This outcome best demonstrates:
Right. The failure was structural, not attitudinal. Many signatories genuinely believed in the pause. The competitive dynamic made unilateral compliance feel too costly.
Why didn't individual labs comply, even if their people believed in the idea? What were they afraid would happen if only they paused?
6. The EU AI Act classifies AI systems by risk level and requires audits for "high-risk" applications. A company in Brazil builds a high-risk AI system and deploys it only in Brazil. The EU AI Act:
Correct. Jurisdiction is territorial. The EU AI Act covers what happens inside EU borders. Powerful regional law cannot solve a global problem when actors operate outside its territory.
Think about what jurisdiction means. The EU is a specific set of countries. Does its law apply to a company operating entirely outside those countries?
7. The UK AI Safety Institute began evaluating AI models before public release. Companies participated voluntarily. What made this governance mechanism meaningful despite lacking legal enforcement?
Right. This is how soft governance works. The standard creates an expectation, and deviation from that expectation has reputational costs. It's weaker than law but stronger than nothing — as long as the norm holds.
If there's no legal enforcement, what pressure exists? Think about what it would mean for a lab to publicly refuse to submit to safety evaluation — and who would notice.
8. A proposal calls for applying the IAEA model to AI — international inspectors, mandatory disclosure, authority to flag dangerous systems. The strongest counterargument is:
Correct. The IAEA's power rests on physical verifiability. AI lacks that property, which means an IAEA-style body would need entirely different verification tools — which don't yet exist at scale.
What specifically allows IAEA inspectors to verify compliance? Is that same thing present in AI development? If not, what would need to replace it?
9. ISO publishes a new AI risk management standard. No law requires compliance. An AI company selling enterprise software decides to comply anyway. The most likely reason is:
Right. Standards create market reference points. When customers, partners, and investors use a standard to evaluate companies, compliance becomes economically rational — even without a law requiring it.
ISO can't sue anyone. So what's the incentive? Think about who uses standards to make decisions and what happens to companies that don't meet them.
10. The lesson describes the Global South as representing about 70% of the world's population but only 12% of participants in AI governance forums. This gap is most concerning because:
Right. Governance reflects the priorities of whoever designed it. Frameworks built without the global majority's input may fail to address their specific concerns — and since these frameworks will persist for decades, the design process right now has long-term consequences for representation and protection.
Think about what "being in the room" means for how standards get written. Whose concerns get prioritized? Whose risks get named? And what happens to the people who weren't consulted?
11. Binding international rules are theoretically better than voluntary ones for addressing the AI race because:
Correct. Binding rules solve the coordination problem by making compliance universal. No one is penalized for following the rule because everyone else has to follow it too. That's the theoretical advantage — the practical challenge is getting the rules agreed to.
Return to the prisoner's dilemma. What makes voluntary cooperation fail? How does a rule that applies equally to everyone change that dynamic?
12. Of the three governance scenarios described in Lesson 4 (Network, UN Agency, Market-Led), which has the highest political feasibility in the near term?
Right. The Network Model is closest to what's actually developing — national institutes, bilateral agreements, shared standards — because it doesn't require the political process of a new international treaty. Its feasibility is also its weakness: it's a partial solution.
Think about what each model requires politically. What does a UN agency need to be created? What does the market-led model assume about how markets work? Which model is building itself already?
13. The lesson argues that institutions don't need perfect enforcement to matter. What is the specific mechanism by which an imperfect institution can still change behavior?
Correct. The practical value is in the standard itself. Once it exists, compliance and deviation are both legible — to investors, journalists, customers, and regulators. That legibility creates pressure even without formal enforcement.
Think about what a standard gives people outside the institution: a reference point. How does a reference point change the conversation between a safety advocate inside a company and their management? How does it help a journalist cover a story?
14. A government announces a new executive order setting AI safety requirements for federal contractors. An AI researcher says this is "just an executive order, not real governance." Are they right?
Right. Executive orders are real governance tools with real effects — but their durability is limited. Biden's AI executive order had real effects until it was rescinded in January 2025. Understanding this distinction helps you evaluate governance announcements accurately.
What did the lesson say about Biden's AI executive order? What happened to it? Does that mean it had no effect? Or does it reveal something about the permanence of executive-branch governance?
15. The module ends with the question: Can we build adequate AI governance before a crisis? Based on the full module, which reasoning best defends the position that pre-crisis governance is necessary even if historically unusual?
This is the strongest argument for urgency. The usual historical pattern — crisis first, governance after — assumes you can recover and learn from the crisis. For certain AI risk categories, the first serious incident may be unrecoverable. That asymmetry changes the calculus of when governance needs to exist.
Think about what makes some risks different from others. If a crisis produces recoverable harm, you can learn and adjust. If a crisis is unrecoverable — catastrophic bioweapon development, for example — what does "learn from the crisis" even mean? How does that change the argument for acting before the crisis?