COMMENTARY | The federal government's responsibility is to establish clear rules before markets become entrenched. #
Thirty years ago, Congress confronted a transformational technology challenge. The Telecommunications Act of 1996 — which spawned MetTel — sought to open telecommunications markets to competition, lower costs and stimulate innovation. Its goals were ambitious and largely bipartisan. Yet the law also provides a cautionary lesson for today's policymakers as they grapple with regulating artificial intelligence.
The challenge facing the federal government is not whether AI should be regulated. The challenge is whether policymakers can create a framework that is unambiguous, encourages innovation, protects national interests and provides market certainty without repeating the costly mistakes that followed telecom reform.
The Telecommunications Act created enormous opportunity. New entrants poured into the market, investors committed hundreds of billions of dollars and entrepreneurs rushed to build the next generation of communications infrastructure. But many of the law's most important provisions lacked specificity. Congress directed incumbent carriers to provide access to competitors yet left key implementation details unresolved.
As a result, regulators, courts and industry participants spent years debating questions that should have been answered from the beginning. What constituted fair access? What technical standards applied? How would disputes be resolved? By the time many of those questions were settled, markets had consolidated, billions of dollars in investment had been lost and much of the competitive landscape Congress intended to create had disappeared.
However, the Act gave birth to a new class of businesses, including MetTel, that heeded the government’s call to open the market to competition using existing telecommunications infrastructure.
For the federal government, the lesson is clear: ambiguity is not neutrality. Ambiguity becomes a policy choice that often shifts decision-making from legislators to regulators, courts and litigants. Artificial intelligence presents a similar challenge, but at a much larger scale.
Unlike telecommunications in 1996, AI is not centered around a single incumbent monopoly. The technology is evolving rapidly across multiple sectors, including healthcare, defense, transportation, financial services, education and critical infrastructure. The United States currently leads much of the world in AI innovation, but that leadership is not guaranteed.
Federal policymakers face a delicate balancing act. Excessive regulation could slow innovation, discourage investment and weaken America's competitive position against strategic rivals. Insufficient oversight, however, could allow critical market bottlenecks to emerge unchecked, creating barriers to competition, reducing innovation and concentrating power in a handful of organizations.
The federal government's responsibility is to establish clear rules before markets become entrenched.
Several potential choke points are already visible. Access to advanced computing resources, control of frontier AI models, availability of high-quality training data and interoperability between systems may ultimately determine which organizations succeed and which are locked out of the market.
If policymakers wait until these issues are fully consolidated, they may find themselves replaying the telecommunications experience: years of litigation, delayed innovation, stranded investments and costly regulatory intervention after the fact. The most important lesson from telecom is that precision matters more than breadth.
Rather than creating broad mandates that require years of interpretation, Congress should establish clear definitions, responsibilities and enforcement mechanisms from the outset. Organizations that control critical AI infrastructure should operate under transparent and predictable rules. Standards for interoperability, particularly in sectors such as healthcare, transportation, public safety and financial services, should be defined before proprietary systems become deeply embedded.
Equally important, enforcement agencies must have the resources and authority necessary to act quickly. Regulatory frameworks are only effective if disputes can be resolved at the speed of technological change. Markets move faster than courts, and technology evolves faster than traditional rulemaking processes. If enforcement lags behind innovation cycles, regulation risks becoming irrelevant.
For the federal government, the stakes extend beyond economics. Artificial intelligence increasingly intersects with national security, cyber resilience, public health, workforce development and America's global competitiveness. Regulatory uncertainty can discourage private-sector investment, while poorly designed regulations can inadvertently strengthen foreign competitors. The objective should not be to control innovation, but to create a stable environment where innovation can flourish while protecting public interests.
The Telecommunications Act demonstrated both the benefits and the limitations of federal policy. It opened markets and created opportunities, but it also left critical questions unanswered, resulting in years of uncertainty and unintended consequences. AI presents an opportunity for policymakers to apply those lessons before similar challenges emerge.
The federal government still has time to shape a competitive, innovative and secure AI ecosystem. Success will depend not on how much regulation is created, but on how clearly it is written and how effectively it is implemented.
History suggests that vague mandates create litigation. Precise rules create markets. As Congress considers the future of AI governance, that may be the most important lesson from the telecommunications reforms of 1996.
As Vice President of MetTel Public Sector, Don Parente is responsible for direct sales and solution architecture for U.S. Federal agencies, states, local and education customers. Don also works with systems integrators to deliver solutions to U.S. government clients.