Who Will Regulate AI?

Artificial intelligence has moved beyond the technology sector and into the foundations of modern business. Companies are using AI across customer service, financial operations, advertising, healthcare, manufacturing, cybersecurity and product development, while investors continue to direct enormous amounts of capital toward the infrastructure supporting the next generation of AI systems. As the economic influence of the technology expands, the question of who should establish its boundaries has become one of the most consequential policy and business debates in the United States.

The debate has become particularly intense in September 2026. Washington is divided over the role of federal government, states are developing their own AI legislation and major technology companies are increasingly calling for clearer national standards. President Donald Trump has continued to resist calls for a significant slowdown in AI development, arguing that the United States cannot afford to lose its technological advantage over China. At the same time, senior figures across the technology industry, including OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei, have been calling for stronger safeguards and greater oversight.

This has created an unusual situation. Some of the companies developing the world's most powerful AI systems are asking governments to establish rules around technologies they are building themselves. OpenAI has publicly supported mandatory national AI safety requirements, including independent assessments, cybersecurity measures and incident reporting for advanced models.

At the heart of the discussion is a difficult economic balance. The United States wants to remain the world's leading AI power, particularly as competition with China intensifies across artificial intelligence, semiconductors, robotics and computing infrastructure. At the same time, policymakers are confronting questions around security, privacy, consumer protection, employment and the potential misuse of increasingly capable systems. Regulation therefore has implications far beyond government policy. It could influence which companies receive investment, how products are developed, how quickly startups can scale and how American businesses compete internationally.

Washington Has Yet To Build A Single AI Rulebook

The United States currently operates without one comprehensive federal law governing artificial intelligence. Instead, companies must navigate a combination of executive actions, existing federal legislation, agency authority and an expanding collection of state-level requirements. This fragmented environment has become increasingly significant as AI moves into more areas of commercial activity.

The federal government's current direction is strongly focused on maintaining American leadership. In June 2026, President Trump signed an executive order titled Promoting Advanced Artificial Intelligence Innovation and Security. The order stated that the United States should avoid overly burdensome regulation while accelerating AI innovation, adoption and investment. It also recognised that advanced AI creates national security considerations requiring coordinated action across federal agencies.

That approach reflects the administration's broader view that excessive regulation could slow American companies at a time when technological competition with China is becoming increasingly important. The White House has positioned AI leadership as both an economic and national security priority, linking technological development with cybersecurity, critical infrastructure and America's broader strategic position.

For businesses, however, innovation policy and regulatory certainty are closely connected. A company investing heavily in an AI product needs to understand not only the market opportunity but also the regulatory environment that could surround that product as it grows. If rules remain fragmented, companies may have to design different compliance strategies for different states, increasing both costs and operational complexity.

That concern has placed the relationship between federal and state governments at the centre of the debate.

The State Versus Federal Government Question

States have increasingly stepped into the regulatory space while Washington has struggled to establish a comprehensive federal framework. California has become one of the most prominent examples, with lawmakers pursuing legislation covering areas such as AI safety, transparency, consumer protection and the use of AI systems.

The growing number of state-level initiatives has created a fundamental question for American businesses. Should individual states retain the ability to establish their own AI rules, or should Washington create a national framework that applies consistently across the country?

The question matters because AI products rarely remain within one state. A software platform developed in California can be used by customers in New York, Texas or Florida within seconds. A model developed by a company in one jurisdiction can provide services to users across the entire country and around the world.

A fragmented regulatory environment could therefore increase compliance costs and create uncertainty for companies trying to scale nationally.

Congress has been considering legislation that would establish stronger federal authority over certain AI risks while potentially limiting state-level rules in specific areas. Senate negotiators were still discussing a bipartisan proposal in September that could impose a duty of care on major AI developers and give the federal government greater authority to prevent the release of models considered dangerously unsafe.

The disagreement is therefore not simply about artificial intelligence. It concerns the structure of American regulation itself and whether a technology that operates across state borders should be governed primarily through national standards or through a collection of state-level rules.

Technology Companies Are Asking For Regulation

The position of the technology industry has added another dimension to the discussion. Some of the largest AI companies are now advocating for mandatory safety requirements, even as policymakers debate how extensive those requirements should become.

OpenAI has called for mandatory national AI safety rules in the United States, arguing that voluntary commitments are insufficient for increasingly capable systems. Its proposal includes capability-based regulation, independent assessments, cybersecurity protocols and incident reporting for advanced AI models.

The company's position is significant because it reflects a growing recognition within the technology industry that regulatory fragmentation can create its own economic problems. A national framework could provide companies with greater certainty and reduce the need to comply with multiple sets of state requirements.

There is, however, another question surrounding this position. When the largest companies participate heavily in designing the regulatory framework, policymakers must consider whether the resulting rules could unintentionally favour established technology companies over smaller competitors.

A requirement that costs millions of dollars to satisfy may be manageable for a global technology company while creating a substantial barrier for an early-stage startup. Regulation can therefore improve safety while simultaneously changing the competitive structure of the market.

The debate is becoming less about whether technology companies support regulation and more about what kind of regulation they support, who writes the rules and how those rules affect competition.

AI Safety Is Becoming A Business Issue

The safety debate has intensified following a series of warnings from researchers and technology executives about increasingly capable AI systems. On September 14, Reuters reported that U.S. senators were considering legislation that would require major AI companies to demonstrate safety measures and potentially allow government auditors to test AI products. The proposal is being discussed on a bipartisan basis, although its prospects remain uncertain.

The political response remains divided. Trump has rejected calls for a significant slowdown, arguing that the United States already has mechanisms capable of policing the industry and warning that slowing development could benefit China.

The technology industry is sending a more cautious signal. Anthropic CEO Dario Amodei has argued for greater oversight of advanced AI development, while OpenAI's Sam Altman and Elon Musk have also supported calls for stronger safety measures.

Microsoft has taken another approach by developing its own proposed code of conduct designed to keep AI systems under human control. The framework includes requirements around correction, shutdown and human oversight and has been opened to public feedback.

These developments show how AI governance is beginning to move beyond abstract policy discussions and into corporate strategy.

The Economics Of Safety

The central economic challenge is straightforward. Advanced AI systems require enormous computing resources, specialised infrastructure and substantial investment. Additional testing, auditing, reporting and cybersecurity requirements could directly affect the cost and speed of development.

For large technology companies, those costs may be manageable. For startups, the situation could be very different.

An emerging company with a promising AI product may possess strong technology but lack the financial resources required to satisfy complex compliance requirements. If regulation becomes too expensive, capital could become increasingly concentrated among a small number of established AI companies.

That would create a paradox. Regulation designed to make AI safer could also make the industry less competitive.

At the same time, the absence of clear standards can create its own problems. Companies may struggle to determine what responsible AI deployment actually requires, investors may face greater uncertainty and consumers may find it difficult to distinguish between systems with different levels of reliability and safety.

The objective should therefore be a regulatory environment that provides meaningful protection while allowing new companies to enter the market.

That balance may become one of the defining questions of the next phase of the AI economy.

The China Factor Is Changing The Calculation

American AI policy cannot be separated from competition with China. The United States and China are competing across AI research, semiconductor production, data infrastructure, robotics and military technology, making technological leadership a central element of economic and national security policy.

This competition has strengthened arguments against regulation that could slow American companies. From this perspective, the United States needs to create an environment where companies can develop and commercialise AI faster than their international competitors.

The Trump administration has repeatedly connected AI leadership with the broader strategic competition with China. The White House's June executive order explicitly framed American AI leadership as a national priority while seeking to accelerate innovation and reduce regulatory barriers.

The opposing argument focuses on resilience and trust. AI systems that become deeply integrated into critical infrastructure, financial markets, government services and cybersecurity need to be reliable and secure. A technology ecosystem that grows rapidly without adequate safeguards could create vulnerabilities that eventually undermine the very competitiveness policymakers are trying to protect.

The disagreement therefore reflects two different interpretations of technological leadership. One prioritises speed and scale. The other prioritises trust, resilience and responsible deployment.

The United States will need to reconcile both if it wants to maintain a durable advantage in AI.

Europe Has Already Taken A Different Path

The European Union has moved further toward comprehensive AI regulation through the AI Act. Unlike the more fragmented American approach, the European framework establishes different obligations according to the risk associated with different AI systems.

Several important provisions became enforceable on August 2, 2026. These include rules covering prohibited AI practices, transparency requirements for certain systems and obligations relating to general-purpose AI models. Additional requirements for certain high-risk AI systems are scheduled to apply at later dates.

This creates an important strategic consideration for global companies.

A business operating in both the United States and Europe may have to design its AI systems around different regulatory philosophies. The European model places greater emphasis on formal risk categories and defined obligations, while the American approach remains more fragmented and focused heavily on innovation, national security and sector-specific oversight.

This difference could eventually influence where AI companies choose to develop products, how they structure compliance teams and how they approach international expansion.

The competition over AI regulation is therefore also becoming a competition over regulatory models.

Regulation Could Reshape Who Wins In AI

Regulation will influence more than safety. It could determine which companies are able to compete.

Large technology companies already possess access to capital, computing infrastructure, specialised talent and extensive legal resources. If advanced AI regulation requires expensive audits, testing procedures and reporting systems, these companies may be better positioned to absorb the additional costs.

Startups could face a different reality.

An emerging company may have the technology to compete but lack the financial resources required to satisfy complex compliance requirements. In that environment, regulation could unintentionally accelerate consolidation within the industry.

Yet the opposite problem is equally significant. Without consistent standards, investors and entrepreneurs may struggle to understand future regulatory exposure. Uncertainty can make funding decisions harder and discourage investment in products whose legal environment remains unclear.

The most effective framework will therefore need to address two objectives at the same time. It must establish meaningful safeguards while preserving enough room for new companies to experiment, compete and introduce new technologies.

AI Regulation Is Becoming A Market-Entry Question

For companies entering the United States, AI regulation is increasingly becoming part of market-entry strategy. A company using AI in customer service, financial services, healthcare, advertising, recruitment or software development may need to understand both the technology and the regulatory environment surrounding its use.

This is particularly relevant for international businesses.

A European company entering the American market may encounter a regulatory environment that differs significantly from the one it already understands. An American company expanding into Europe faces the opposite challenge.

As AI becomes embedded into more commercial products, regulatory knowledge will increasingly become part of international expansion planning. The question of where a company can deploy an AI system may eventually depend on more than technical capability and customer demand. Compliance architecture could become part of the product itself.

This makes AI governance relevant to founders, investors and executives far beyond the technology industry.

Who Should Regulate AI?

There may never be a single institution responsible for every aspect of artificial intelligence.

Federal agencies already possess authority over areas including consumer protection, financial markets, healthcare, cybersecurity and national security. The federal government is also using executive action to coordinate AI policy across agencies and accelerate adoption in areas such as national security and critical infrastructure.

States will continue to influence areas where they retain regulatory authority, while international frameworks will shape companies operating across borders.

The more realistic question is therefore how these different layers of governance should interact.

Federal government can establish national standards. States can address specific local concerns. Agencies can regulate AI within their existing areas of authority. International frameworks can establish requirements affecting companies operating across borders.

The challenge is creating enough coordination to prevent regulatory fragmentation while preserving the ability to respond to emerging risks.

The Business World Needs Predictability

Businesses can adapt to strict regulation more easily than unpredictable regulation.

A company can build its product around clear requirements, budget for compliance and establish internal processes when it understands what the rules are. Constantly changing requirements create a different problem because businesses cannot easily determine whether today's investment will remain viable tomorrow.

For investors, this uncertainty can influence valuations and capital allocation. For entrepreneurs, it can influence whether a product is launched in a particular market. For established companies, it can influence technology procurement and long-term strategy.

Regulatory certainty is therefore becoming an economic asset.

The United States has historically benefited from an environment where entrepreneurs could build companies at enormous scale. Maintaining that advantage in the AI era will require policymakers to create rules that protect the public without making innovation accessible only to the largest corporations.

The Next Phase Will Be About Governance

The American AI debate is entering a more consequential phase. Washington is discussing federal standards, states are advancing their own legislation, technology companies are calling for certain forms of mandatory oversight and international regulators are establishing frameworks that will influence global businesses.

The outcome will affect far more than AI developers.

It will influence financial institutions adopting AI, retailers using automated customer systems, manufacturers deploying robotics, advertisers relying on generative technology and entrepreneurs building new companies around AI-enabled products.

The regulatory question is therefore becoming a business question.

The United States now faces a choice between allowing a largely fragmented system to develop organically or building a stronger national framework capable of creating greater consistency across the country. Both approaches carry risks.

A highly centralised framework could slow innovation and increase barriers for startups. A fragmented system could increase compliance costs and create uncertainty for companies operating nationwide. A balanced model will need to recognise both concerns.

The answer may ultimately come through a combination of federal standards, specialised agency oversight and carefully defined state authority.

What matters most for the business community is that the rules become sufficiently clear for companies to plan around them.

Artificial intelligence is moving too quickly for regulation to remain an afterthought. The technology is already influencing how companies operate, how consumers interact with businesses and how investors evaluate future growth.

The United States has an opportunity to shape the next generation of the AI economy, yet technological leadership will depend on more than building the most powerful models or operating the largest data centres. It will also depend on creating an environment in which innovation can scale with public confidence.

The central question is no longer whether AI will be regulated. The direction of travel suggests that regulation is becoming increasingly difficult to avoid as the technology moves deeper into the economy.

The more consequential question is who will write the rules, whose interests those rules will represent and whether the United States can create a regulatory framework that protects society without weakening the entrepreneurial system that made its technology industry so powerful.

That decision will shape the competitive landscape of AI for years to come.