Why AI Governance and Trust Are Becoming Institutional Challenges

The AI era is creating a new kind of governance challenge: how institutions maintain public trust when information itself becomes increasingly synthetic, automated, and difficult to verify. Deepfakes, synthetic media, and algorithm-driven communication are beginning to reshape how societies interpret credibility, authority, and truth. At the same time, many discussions around AI governance remain trapped at the level of abstract ethics rather than operational accountability.

Most of the conversation around AI governance is still abstract – centered on principles, ethics, and broad guidelines. What’s missing is how this works in practice.

As artificial intelligence continues to shape societies and institutions, it raises a broader question: how can democratic values like accountability, transparency, and public trust be protected in an increasingly AI-driven world?

Michael S. Dukakis and Nguyen Anh Tuan argue that preserving public trust in the AI era will require more than regulation alone; it will demand stronger institutional accountability, clearer governance frameworks, and a renewed focus on democratic values in technological systems. As co-authors of ‘America at 250: A Beacon for the AI Age’, they reflect on the evolving relationship between technology, leadership, and society.

The challenge of governing AI democratically

As AI increasingly shapes public discourse, decision-making, and access to information, questions around trust, accountability, and democratic leadership have become more urgent. This section examines how leaders/institutions can preserve transparency, human judgment, and democratic values in a rapidly evolving AI-driven society.

Michael Dukakis (Former Governor of Massachusetts and the Co-Founder and Chair of the Boston Global Forum)   

As someone who has spent years in government leadership, how do you define “democratic leadership” in an AI-driven world, where influence is increasingly shaped by algorithms rather than institutions?

Dukakis: As America approaches its 250th anniversary, democratic leadership must mean more than holding office or winning elections. It must mean protecting the dignity, rights, and voice of every citizen in a world increasingly shaped by artificial intelligence.

In the past, democratic leaders mainly exercised influence through institutions such as legislatures, courts, a free press, elections, universities, and civic organizations. These institutions remain essential. But today, algorithms influence what people see, what they believe, how they communicate, and sometimes even how they understand democracy itself.

That means democratic leadership in the AI Age requires a new responsibility: to ensure that technology serves people, not the other way around.

Algorithms must not become invisible powers that manipulate citizens, divide communities, or weaken public trust. They must be transparent, accountable, and guided by human values. We need leaders who understand that innovation without responsibility can endanger freedom, while innovation guided by democracy can strengthen it.

For America at 250, this is a historic test. The United States should not just lead by building the most powerful technologies. But also by demonstrating how artificial intelligence can be governed with trust, fairness, compassion, and respect for human dignity.

To me, democratic leadership in an AI-driven world means keeping human beings in command. It means defending truth against disinformation, protecting privacy, expanding opportunity, and ensuring that AI strengthens democratic institutions rather than replacing or undermining them.

America’s greatest contribution has never been power alone. It has been the belief that free people, guided by shared values and responsible institutions, can build a better future. In the AI Age, that mission is more important than ever.

You’ve spent decades in public service. What lessons from traditional governance do you think still hold, and where do they fall short when applied to AI oversight?

Dukakis: The first lesson from traditional governance is that public trust matters. No government can function well without trust. And no technology should be deployed in ways that weaken that trust.

From my years in public service, I learned that good governance depends on accountability, transparency, fairness, and service to the public interest. These principles still remain relevant today. They are more important than ever in the AI Age.

AI oversight must begin with the same democratic questions we have always asked: Who is responsible? Who benefits? Who may be harmed? What rights must be protected? How can citizens seek redress when something goes wrong?

Those lessons still hold.

But traditional governance also has limits. Government often moves slowly, while AI develops at extraordinary speed. Laws and regulations can take years to design and pass, while AI systems can be deployed globally in months or even weeks.

Traditional oversight was designed for institutions and human decision-making. Not for complex, fast-changing, and often difficult-to-explain AI systems.

That is where traditional governance falls short.

We need new tools for continuous oversight: independent auditing, testing before deployment, monitoring after deployment, clear standards for high-impact AI systems, and strong protections for privacy, security, and human dignity. We also need what the Boston Global Forum and AIWS have called Trust Infrastructure – a framework that can measure, verify, and sustain trust in AI systems over time.

But the foundation remains the same: technology must be accountable to democracy. AI must never become an excuse to avoid responsibility. Human beings must remain in command, especially in decisions affecting rights, health, security, justice, and the future of society.

So the lesson is this: we should bring the best values of traditional governance into the AI Age, but we must also build new institutions, standards, and safeguards equal to the power of this technology.

Your book, ‘America at 250: A Beacon for the AI Age’, argues for aligning innovation with democratic values. In practical terms, how should governments intervene without stifling technological progress?

Dukakis: Governments should not try to stop innovation. They should guide it so that innovation strengthens society rather than harms it.

In practical terms, that means setting clear rules for areas where AI can affect people’s rights, safety, privacy, health, jobs, education, security, and democratic participation. The goal is not to regulate every experiment or every new idea. The goal is to ensure powerful AI systems meet standards of transparency, accountability, fairness, and human control before organizations deploy them in real-world settings.

Good government can actually accelerate responsible innovation. When citizens, businesses, and institutions trust a technology, adoption becomes stronger and more sustainable. Trust is not the enemy of progress. Trust is the foundation of progress.

There are several practical steps governments should take.

• First, high-impact AI systems should undergo testing and independent audits before organizations deploy them widely

• Second, people should have a clear understanding of when AI is being used. Especially in decisions that directly affect their lives.

• Third, governments and companies must protect privacy and data dignity. So that citizens are not treated merely as data sources.

Fourth, require clear responsibility. When an AI system causes harm, there must be a human institution accountable for that harm.

Fifth, support innovation through research funding, public-private partnerships, education, and open standards, so that democratic societies do not fall behind.

This is why the book emphasizes the need for Trust Infrastructure. We need systems that can measure, verify, and continuously monitor whether AI is serving human dignity and democratic values.

The right approach is not heavy-handed control, and it is not a free-for-all. It is responsible leadership. Government should create the guardrails, standards, and public trust that allow innovation to flourish in the right direction.

At America’s 250th anniversary, the United States has an opportunity to show that the most advanced technology can also be the most humane, the most trusted, and the most faithful to democratic ideals.

Public trust in institutions – governments and public systems- has been declining. Do you believe AI governance can help restore that trust, or does it risk deepening skepticism?

Dukakis: AI governance can either help restore public trust or deepen public skepticism. It depends entirely on how we build it.

Hidden, manipulative, biased, or unaccountable uses of AI will only deepen the trust crisis. People will feel that decisions affecting their lives are being made by systems they cannot understand, cannot question, and cannot appeal. That would be very dangerous for democracy.

Effective AI governance can also help rebuild that trust.

Good AI governance should make institutions more transparent, more responsive, and more accountable. It can help governments deliver services faster, detect problems earlier, reduce waste, expand access to education and health, and give citizens better ways to interact with public systems. But these benefits will only matter if people believe the systems are fair, secure, and under human control.

That is why trust must be designed into AI from the beginning. We need clear standards, independent audits, public reporting, human oversight, privacy protection, and meaningful redress when mistakes are made. Citizens should know when AI is being used, why it is being used, and who is responsible for the outcome.

This is also why the Boston Global Forum and AIWS emphasize Trust Infrastructure. Trust cannot be built by speeches alone. It must be measured, verified, and maintained over time.

For democracy, the central question is not simply whether AI is powerful. The question is whether AI makes public institutions more worthy of trust.

If AI helps governments serve people with honesty, fairness, and compassion, it can become a tool for renewing democratic confidence. But if AI is used to avoid responsibility or concentrate power, it will deepen skepticism.

So my answer is: yes, AI governance can help restore trust – but only if it is rooted in human dignity, transparency, accountability, and the firm principle that human beings must remain in command.

In your view, how should organizations in the U.S. balance national interests with the need for global cooperation on AI standards, especially when geopolitical tensions are high?

The United States must protect its national interests, but it should never confuse national strength with isolation.

In the AI Age, no country can build a safe and trustworthy future alone. AI systems cross borders. Data flows across societies. Cyber threats, disinformation, autonomous systems, and economic disruption do not stop at national boundaries. That means American leadership must combine strength at home with cooperation abroad.

For organizations in the United States, the first responsibility is clear: protect national security, safeguard citizens, defend democratic institutions, and ensure that AI serves the public interest. America must lead in innovation, research, talent, infrastructure, and responsible deployment.

But leadership also requires building common standards with trusted partners. The U.S. should work closely with democracies and responsible nations to establish shared principles for AI safety, transparency, privacy, accountability, human control, and the prevention of misuse.

This is especially important when geopolitical tensions are high. We cannot be naïve. Some regimes may use AI for surveillance, repression, manipulation, or military advantage. Democracies must be prepared to defend themselves. But the answer is not to abandon cooperation.

The answer is to build alliances based on trust.

We need what I would call a democratic framework for AI cooperation -a network of nations, universities, companies, civil society organizations, and public institutions committed to common standards and responsible innovation.

This is where Trust Infrastructure becomes essential. If countries and organizations can measure, verify, and certify trustworthy AI systems, then cooperation becomes more practical and more credible. Trust should not be merely a statement of goodwill. It must be built into standards, audits, accountability, and shared commitments.

So the balance is this: America must be strong, competitive, and secure. But it must also be generous, principled, and collaborative.

At its best, the United States leads not by domination, but by example – by showing that technological power can be guided by democratic values, human dignity, and a responsibility to humanity. That is the kind of leadership the world needs in the AI Age.

What is one concrete change you hope to see in the next five years?

Dukakis: One concrete change I hope to see in the next five years is the establishment of trusted, independent standards for high-impact AI systems.

By that I mean: before AI is used in areas such as public services, health care, education, finance, elections, security, or justice, it should be tested, audited, and certified according to clear public standards. Citizens should know when AI is being used, what safeguards are in place, and who is accountable if something goes wrong.

This would be a major step toward restoring public trust.

We do not need to wait decades to do this. In the next five years, governments, universities, companies, and civic organizations can work together to build practical Trust Infrastructure – systems for transparency, accountability, independent evaluation, and continuous monitoring.

If we can create a world where powerful AI systems are not only innovative but also trustworthy, explainable, secure, and guided by human dignity, that would be a historic achievement.

For America at 250, that should be part of our national mission: to show that democracy can lead the AI Age not only with technology, but with values, responsibility, and trust.

Responsible AI in Practice: Trust, Transparency, and Global Collaboration

As AI technologies continue to expand across industries and national boundaries, building trust cannot be addressed solely by individual organizations or governments. Let’s how cross-border collaboration, shared standards, and institutional accountability can help create more trustworthy and interoperable AI ecosystems.

Nguyen Anh Tuan (Co-Founder, Co-Chair, and CEO of the Boston Global Forum and AIWS)

What are the biggest barriers leaders face when trying to operationalize “trust” in AI systems?

Tuan: The biggest barrier is that many leaders still treat trust as a slogan rather than an operating system.

In the AI Age, trust cannot be built only through principles, speeches, or ethical declarations. Trust must be translated into standards, data, metrics, audits, accountability, and continuous monitoring. It must be operationalized.

• The first barrier is measurement. Leaders often agree that AI should be trustworthy, but they lack clear ways to assess whether a system is safe, transparent, fair, secure, accountable, or aligned with human values. Without measurement, trust remains abstract.

• The second barrier is accountability. AI systems are often complex, involving developers, deployers, data providers, vendors, regulators, and users. When harm occurs, responsibility can become unclear. Trust requires a clear chain of accountability.

• The third barrier is speed. AI develops faster than traditional institutions can respond. Governments, universities, and companies often rely on governance models designed for slower technologies. AI requires continuous oversight, not occasional review.

• The fourth barrier is data quality and transparency. Many AI systems depend on data that may be incomplete, biased, insecure, or difficult to trace. Without trustworthy data, we cannot have trustworthy AI.

• The fifth barrier is geopolitical fragmentation. Different nations and companies are developing AI for different interests, standards, and values. Without shared trust standards among democratic societies and responsible partners, AI can deepen division and conflict.

But these barriers are not reasons for delay. There are reasons for leadership.

I am encouraged that a growing number of visionary leaders in government, business, academia, and civil society are ready to work with the Boston Global Forum and AIWS to implement AIWS Trust Infrastructure in practical settings. They understand that trustworthy AI must move beyond discussion and become part of real governance, real enterprise practice, and real public systems.

BGF-AIWS stands ready to support these leaders with concrete solutions: trust standards, assessment models, AIWS Trust Rating, AIWS Trust Index, independent auditing frameworks, information provenance, incident reporting, human-in-command safeguards, and implementation roadmaps adapted to different national, regional, and organizational contexts.

This is why I believe we need AIWS Trust Infrastructure. Trust must become a practical architecture: standards, testing, independent auditing, trust ratings, trust indexes, incident reporting, and human-in-command safeguards.

The question is no longer whether we support trustworthy AI. Everyone says they do. The real question is whether we can build the systems that make trust measurable, verifiable, and enforceable.

That is the mission of AIWS Trust Architecture: to move from aspiration to implementation, from values to infrastructure, and from trust as a promise to trust as a living system for the AI Age.

How do you ensure that frameworks like AI World Society (AIWS) remain adaptable as AI technologies evolve rapidly, especially with generative AI and synthetic media?

Tuan: AIWS must remain adaptable because AI itself is not a fixed technology. It is a living force, evolving from machine learning to generative AI, agentic AI, synthetic media, frontier models, autonomous systems, and future forms we cannot yet fully imagine.

To keep AIWS relevant, we must design it not as a rigid rulebook, but as a living architecture.

AIWS always looks ahead. We welcome new technologies, follow the latest advances in AI, and study emerging breakthroughs not only as risks to be managed, but also as advantages and solutions for the problems society urgently needs to solve.

Rapid technological progress should not make governance passive or fearful. It should inspire better standards, better tools, and better institutions.

• The first principle is continuous learning. AIWS must observe new technological developments, study emerging risks, identify new opportunities, and update its standards accordingly. A framework for the AI Age cannot be revised once every ten years. It must be capable of learning in real time.

• The second principle is modularity. AIWS should have core values that do not change — human dignity, trust, accountability, transparency, compassion, and human-in-command — but its tools, metrics, and implementation methods must be flexible. The values are stable; the mechanisms must evolve.

• The third principle is continuous assurance. With generative AI and synthetic media, the danger is not only that machines make mistakes, but that they can reshape public perception, create false realities, and undermine trust in democracy. That is why AIWS Trust Infrastructure must include monitoring, provenance, watermarking, content authentication, independent audits, and rapid response to information attacks.

• The fourth principle is multi-stakeholder governance. AIWS cannot be built by one government, one company, or one academic institution alone. It must be shaped by leaders in government, technology, civil society, media, education, security, and culture. This diversity makes the framework stronger and more adaptive.

• The fifth principle is human-in-command. As AI systems become more autonomous, society must be very clear about which decisions should remain under human responsibility. AI may assist, analyze, and recommend, but in high-stakes areas, humans must remain accountable.

Generative AI and synthetic media underscore the need for AIWS.

We are entering an age when seeing is no longer believing, and when information itself can be manipulated at scale. Therefore, trust must become infrastructure – measurable, verifiable, and continuously protected.

AIWS will remain adaptable by holding firm to enduring human values while constantly updating the standards, technologies, and institutions needed to defend and advance those values in a changing AI world.

Our approach is not to resist the future, but to guide it. AIWS seeks to transform the newest achievements of AI into trusted solutions for humanity – solutions that strengthen democracy, protect dignity, expand creativity, and build a wiser civilization in the AI Age.

What role should the private sector play in building and maintaining AI trust infrastructure, and where should accountability ultimately lie?

Tuan: The private sector has a vital role to play, because much of the AI revolution is being built by companies – from frontier models and cloud infrastructure to applications in health, finance, education, media, security, and government services.

But the private sector should not be allowed to define trust by itself.

Companies must help build AI trust infrastructure in very practical ways. They should implement safety-by-design, privacy-by-design, transparency, provenance, cybersecurity, model evaluation, red-teaming, incident reporting, and continuous monitoring. They should also provide evidence that their systems meet clear trust standards before those systems are deployed in high-impact areas.

The private sector also has the technical capacity, data, engineering talent, and innovation speed that governments often lack. Without companies, AI trust infrastructure will remain theoretical. But without public accountability, private-sector trust systems can become self-certification – and that is not enough.

Trust must be verified, not merely claimed.

That is why AIWS Trust Infrastructure should be built through cooperation among companies, governments, universities, independent auditors, think tanks, civil society organizations, and international partners. Companies can design and operate trustworthy systems, but independent institutions must test, audit, rate, and monitor them.

Think tanks and civil society organizations have a particularly important role. They help ensure that AI trust infrastructure is not distorted by short-term commercial incentives or by the narrow interests of any single company or industry. They bring public-interest perspectives, ethical reflection, independent research, civic accountability, and the voice of society into the governance of AI.

This is not a position against business. On the contrary, responsible companies need trusted institutions around them. A healthy AI ecosystem requires both innovation and independent verification. The private sector brings speed and capability. Civil society, universities, think tanks, and independent bodies bring legitimacy, scrutiny, and public trust.

At the same time, we must also solve an important practical question: how can civil society organizations, think tanks, universities, and independent experts be supported when they contribute to building and operating AIWS Trust Infrastructure? Their participation requires resources, recognition, incentives, and sustainable models.

In America at 250: A Beacon for the AI Age, I discuss the idea of AIWS Rewards and a new economy of social contribution. In the AI Age, people and organizations that contribute to public trust, ethical innovation, transparency, civic knowledge, and the common good should be recognized and rewarded. Trust infrastructure should not depend only on government budgets or corporate funding. It should also create mechanisms that honor and sustain meaningful contributions to society.

This can include recognition for organizations that contribute to standards, audits, education, data integrity, information provenance, incident reporting, civic resilience, and responsible AI adoption. In this way, AIWS Trust Infrastructure can become not only a governance framework, but also a new ecosystem where social contribution has real value.

Accountability ultimately must lie with human institutions and human leaders.

An AI system cannot be morally or legally responsible by itself. The company that builds it, the organization that deploys it, and the public authority that permits or regulates its use must each carry clear responsibility. In high-stakes areas – such as public services, health, education, finance, elections, security, and justice – there must always be a human-in-command.

The private sector should be a builder and partner of trust infrastructure. Government should set the public-interest guardrails. Independent bodies should verify compliance. Think tanks and civil society should help protect the integrity and legitimacy of the system. Citizens should have transparency, rights, and redress.

In the AI Age, trust cannot be outsourced. It must be shared, measured, verified, and sustained. The private sector can help build the architecture, but accountability must remain with responsible human leadership, democratic institutions, and a broader ecosystem committed to the common good.

In your view, what concrete mechanisms can prevent large-scale misinformation without infringing on free expression?

Tuan: The challenge is to protect society from large-scale misinformation without creating a system of censorship. In the AI Age, this is one of the most delicate and important tasks for democracy.

Free expression must remain a foundational value. People must have the right to express opinions, criticize power, debate public issues, create art, practice satire, and participate freely in civic life. But in the Internet and AI Age, free expression also requires a more mature understanding of responsibility. When information, images, audio, video, and data can be created and distributed at a massive scale, society must also demand higher standards of trust, accuracy, authenticity, and accountability.

Freedom of expression should not mean freedom to secretly manipulate society.

There is fabricated evidence, impersonation, deepfakes, bot networks, or synthetic media designed to deceive the public. The goal is not to restrict legitimate speech. The goal is to ensure that information entering the public sphere can be understood, verified, and trusted.

• The first mechanism is provenance. People should know where information comes from, whether an image, video, audio, or text was created by a human, edited by AI, or generated by AI. This does not silence speech. It gives citizens context.
• The second mechanism is content authentication. Trusted public communications — from governments, election authorities, health agencies, universities, and responsible media — should carry verifiable digital signatures. Citizens should be able to confirm whether a message is authentic.
• The third mechanism is labeling and disclosure, especially for synthetic media. AI-generated political ads, deepfake videos, impersonations, or manipulated public statements should be clearly identified. Free expression does not require deception.
• The fourth mechanism is independent fact-checking and rapid response, especially during elections, crises, wars, pandemics, or major public emergencies. The response should focus on correcting false claims, exposing manipulation, and providing reliable information — not suppressing legitimate debate.
• The fifth mechanism is algorithmic transparency and accountability. Platforms should be required to assess whether their recommendation systems amplify harmful misinformation at scale. The problem is not merely that false speech exists; the danger is when algorithms push it to millions of people because it increases engagement.
• The sixth mechanism is media literacy and civic education. Citizens must be equipped to recognize manipulation, deepfakes, bot networks, and emotional propaganda. A democratic society cannot depend only on regulation; it must strengthen the judgment of its people.

This is why AIWS proposes an Information Trust Infrastructure.

Its purpose is not to decide what people are allowed to think. Its purpose is to help people distinguish authentic information from manipulated information, credible sources from artificial deception, and public truth from engineered chaos.

BGF-AIWS has an important principle: we should not be paralyzed by risks and challenges. Every age has its own dangers, and every major technological transformation creates new vulnerabilities. But those challenges should become a lever for society to become more mature, more responsible, and more capable.

In the AI Age, we should use the newest achievements of AI itself to address and master the problems that AI can create. AI can help detect synthetic media, trace provenance, identify coordinated manipulation, verify authenticity, monitor information attacks, support fact-checking, and strengthen civic resilience. The answer to AI-generated misinformation is not fear of AI. The answer is better AI, governed by trust, transparency, human responsibility, and democratic values.

Free expression must remain protected. People have the right to opinions, criticism, satire, and political disagreement. But no one should have the right to secretly use AI to impersonate leaders, fabricate evidence, manipulate elections, or destroy public trust at scale.

The democratic answer is not censorship. The answer is transparency, provenance, authentication, accountability, civic education, and trust infrastructure — so that truth can compete fairly in the public sphere, and so that society can become wiser and more resilient in the face of new challenges.

Given the global nature of AI, how do you envision cross-border collaboration working in practice – particularly between democratic and non-democratic systems?

Tuan: Cross-border collaboration on AI must be practical, principled, and realistic.

AI is global by nature. Models, data, cyber risks, synthetic media, supply chains, talent, and standards all cross national borders. No nation can govern AI alone. But cooperation cannot be naïve, especially when democratic and non-democratic systems may have very different views of human rights, privacy, surveillance, and the role of the state.

I believe collaboration should work on two levels.

First, democratic societies and trusted partners should build a strong common foundation. They should agree on shared standards for AI safety, transparency, accountability, privacy, human-in-command, and protection against misuse. This democratic trust network can become the core of a new AIWS Trusted Order – a framework where trustworthy AI is measured, audited, certified, and continuously monitored.

Second, democracies should still engage non-democratic systems where cooperation is necessary for global safety. There are areas where the world needs minimum common rules: preventing catastrophic AI misuse, reducing cyber escalation, managing autonomous weapons risks, protecting critical infrastructure, and responding to large-scale disinformation or deepfake attacks.

But such cooperation must be based on verification, not blind trust.

That is why AIWS Trust Infrastructure is important.

It provides a way to move from political statements to measurable commitments. Countries and organizations can be evaluated according to standards, evidence, audits, incident reporting, and compliance mechanisms. Trust must be earned through behavior, not claimed through declarations.

In practice, we should begin with implementation among trusted partners, learn from real-world deployment, improve the framework, and then expand. The first phase can begin with the United States, Japan, Vietnam, and Southeast Asia, together with other democratic and responsible partners. These are places where strategic trust, innovation capacity, and urgent social needs can come together to create practical models for AIWS Trust Infrastructure.

The America at 250 Conference at Harvard University called for this kind of action through its Declaration on founding AIWS Trust Infrastructure and AIWS Information Trust Infrastructure for the AI Age. The Declaration was not only symbolic. It was a call to move from vision to implementation – to build working standards, trust ratings, information provenance systems, auditing mechanisms, and institutional partnerships.

This is also the spirit of my book with Governor Michael Dukakis, America at 250: A Beacon for the AI Age. The book is not only about ideas. It is about solutions that can be implemented immediately. It proposes pathways to turn democratic values, human dignity, and trust into operating systems for governance, business, technology, and international cooperation.

In practice, I envision regional and global networks of governments, companies, universities, auditors, media organizations, think tanks, and civil society working together on common protocols: AI trust ratings, information provenance, emergency response to synthetic media attacks, safety testing for high-impact AI, and shared learning from incidents.

We should not wait for a perfect global consensus before beginning.

We should start with committed partners, build credible models, test them in real environments, adjust them based on evidence, and then scale them internationally. This is how trust infrastructure can become real.

Democracies should lead this process with openness and confidence. They should not isolate themselves from the world, but they must also not compromise the core values of human dignity, freedom, privacy, and accountability.

The goal is not to impose one system on everyone. The goal is to create a trusted architecture where cooperation is possible, risks are reduced, and humanity remains in command of AI. In the AI Age, peace and security will depend not only on power, but on trust that can be measured, verified, implemented, and sustained across borders.

If you had the chance to establish one global rule for AI that all countries must follow, what would it be?

Tuan: The one global rule I would establish is this:

AI must always remain under human responsibility and must never be allowed to override human dignity, human rights, or human life.

This is the principle of human-in-command.

As humanity enters the AI Age, we are stepping into a new civilizational transformation — a new revolution that will reshape the way we live, work, govern, create, and relate to one another. In such a time, the question is not only what AI can do. The deeper question is what kind of world we want AI to help us build.

For AIWS, the answer is clear. We want to help build a world in which every person is respected, where each human being has the opportunity to develop his or her full potential, where opportunities are fairer and more equal, where people can live in peace and safety, where they are cared for and valued, where creativity is encouraged, and where society is guided toward noble values.

Any global rule for AI should be judged by whether it helps advance that human future.

AI can assist human beings. It can analyze, predict, recommend, create, and help solve complex problems. But AI must not become the final authority over decisions that affect human dignity, freedom, safety, justice, or the future of society.

This rule would apply especially to high-stakes areas: nuclear weapons, autonomous weapons, public security, health care, justice, elections, education, financial systems, and critical infrastructure. In these areas, there must always be identifiable human responsibility, clear accountability, and meaningful redress.

The world needs many AI rules – for transparency, privacy, safety, cybersecurity, provenance, and fairness. But all of them should rest on one foundation:

AI must serve humanity, not control humanity.

This global rule would also require practical mechanisms: independent audits, safety testing, incident reporting, human oversight, stop-switches, and international standards for trustworthy AI. It is not enough to say humans are in command. We must build systems that make that principle real.

For AIWS, this is at the heart of Trust Infrastructure. Trust begins when people know that no machine, no algorithm, and no hidden system can take away human responsibility.

The AI Age will test every civilization. The most important question is not whether AI becomes more intelligent. The most important question is whether humanity remains wise, responsible, compassionate, and in command. So my global rule would be simple, but profound:

AI must always be accountable to human dignity, human responsibility, and the common good of humanity – and it must help build a world of peace, safety, creativity, fairness, and noble human flourishing.

As AI adoption accelerates across the Indo-Pacific and the Global South, how important will international collaboration be in creating trusted AI infrastructure, interoperable standards, and governance models that organizations can realistically implement at scale?

Tuan: International collaboration will be essential. One nation, one company, or one region alone cannot build an AI system that is trustworthy. AI systems move across borders. Data, platforms, models, cyber risks, synthetic media, and supply chains are all global. International cooperation is very much needed to build a trusted AI infrastructure.

This is especially important for the Indo-Pacific and the Global South. These regions are not merely adopters of AI. They will be major arenas where AI shapes economic development, education, health care, public services, finance, agriculture, media, and national security. Without practical, interoperable, trusted AI standards, many organizations risk falling behind or will be pushed to adopt systems they cannot fully understand, audit, or control.

The challenge is to make AI governance implementable.

Many countries and organizations support the idea of trustworthy AI, but they need concrete tools: standards, assessment methods, audit models, trust ratings, incident reporting, provenance systems, cybersecurity safeguards, and human-in-command requirements. Governance must be translated into mechanisms that a ministry, a company, a university, a bank, a hospital, or a media organization can actually use.

Hence, AIWS Trust Infrastructure goes beyond being just an idea and focuses on creating a practical framework. It can help create common standards while still allowing flexibility for different national contexts and levels of development. The goal is not to impose one model on every society. The goal is to build a trusted framework that enables interoperability, accountability, and cooperation.

In practice, I believe the Indo-Pacific can become a pioneering region for trusted AI. The United States, Japan, India, Vietnam, ASEAN countries, Europe, and other democratic and responsible partners can work together to build shared standards for AI safety, transparency, privacy, provenance, information integrity, and human responsibility. These standards must protect society while also remaining practical enough for large-scale adoption.

For the Global South, this is also a question of fairness.

Trusted AI infrastructure should not be a privilege of wealthy nations alone. If AI is to serve humanity, then emerging economies must have access to trustworthy systems, training, evaluation tools, and governance models that help them develop safely and confidently.

International collaboration can also prevent fragmentation. Without shared standards, the world may divide into incompatible AI systems governed by different values, different levels of transparency, and different approaches to human rights. That would increase risk and reduce trust. Interoperable standards can help create a more stable and responsible AI ecosystem.

Ultimately, the future of AI governance will depend on whether we can move from declarations to implementation. We need cooperation that produces real standards, real audits, real trust ratings, real monitoring, and real accountability.

For AIWS, this means building a global framework where AI remains innovative, secure, human-centered, and trusted across borders. International collaboration is not optional. It is the foundation for building AI that serves the common good of humanity.

Many organizations are realizing that trusted AI requires more than technical safeguards alone. How does AIWS Lumina complement AIWS Trust Infrastructure in helping enterprises build a stronger culture of responsible AI adoption, transparency, and human-centered innovation?

Tuan: Technical safeguards alone cannot build trusted AI. Standards, audits, cybersecurity, data governance, provenance, and monitoring are essential. But they are not enough. At the deepest level, trusted AI also depends on culture – the values, habits, responsibilities, and human spirit of the organizations that design, deploy, and use AI.

This is where AIWS Lumina complements AIWS Trust Infrastructure.

AIWS Trust Infrastructure provides the architecture of trust: standards, metrics, audits, trust ratings, accountability mechanisms, human-in-command safeguards, and continuous monitoring. It helps organizations answer practical questions: Is this AI system safe, transparent, and accountable? Can privacy be protected? Is it possible to audit the system? Who is responsible if harm occurs?

AIWS Lumina adds the cultural and human dimension. It helps organizations ask deeper questions: What kind of future are we building? Does this technology elevate human dignity? Can it strengthen trust between people? Does it encourage creativity, compassion, responsibility, and nobility? Does it help human beings flourish?

For enterprises, this is very important.

Responsible AI adoption is not a compliance issue, but it is also a leadership issue.

A company may have good policies on paper, but if its culture rewards speed without reflection, growth without responsibility, or automation without human care, then trust will remain fragile.

AIWS Lumina encourages enterprises to build a culture guided by Love, Creativity, and Nobility. Love means technology should serve people with respect and compassion. Creativity means AI should expand human imagination and innovation, not reduce human beings to data or replace human judgment. Nobility means organizations should act with responsibility, transparency, and a sense of higher purpose. Together, AIWS Trust Infrastructure and AIWS Lumina create a fuller model for trusted AI

Trust Infrastructure gives enterprises the mechanisms to measure and verify trust. Lumina gives them the cultural foundation to live trust in daily decisions.

In practical terms, this means enterprises should not treat responsible AI as the job of one compliance team or one technology department. It must become part of leadership training, product design, employee culture, customer communication, risk management, and public accountability. Engineers, executives, managers, and users should all understand that AI must serve human dignity and the common good.

AIWS Lumina can also help enterprises communicate trust more clearly. Transparency is not only about disclosure. It is also about sincerity, respect, and the willingness to explain decisions in ways people can understand. Human-centered innovation is not only about usability. It is about building technologies that strengthen human agency, human creativity, and human responsibility.

In the AI Age, the most successful organizations will not be those that adopt AI the fastest, but those that adopt AI wisely. They will combine technical excellence with ethical responsibility, operational trust with cultural depth, and innovation with humanity.

That is the role of AIWS Lumina: to ensure that AIWS Trust Infrastructure is not only a system of safeguards, but also a living culture of wisdom, dignity, creativity, and trust.

Key takeaway:

This conversation highlights a clear message: the future of AI must be guided not only by technological progress, but also by responsible AI practices rooted in democratic values, global cooperation, and trust-centered leadership. Without strong AI governance and trust, even the most advanced AI systems risk creating greater uncertainty, misuse, and accountability challenges.