Ajit Pai: The Big Lesson from Barcelona is AI and Wireless Are Converging

Last week, the global wireless industry convened at Mobile World Congress in Barcelona, Spain to explore the future of connectivity. While broadband and 5G were still very much at the forefront of those conversations, one idea stood out above all the rest: leadership in innovation increasingly relies on the convergence of artificial intelligence and wireless.

Reminders were replete on the show floor: humanoid robots interacting with attendees, AI-powered assistants navigating simulated hospital wards, smart-factory demonstrations coordinating over wireless networks, and more.

The message was unmistakable: AI is not a passive abstraction sitting in a data center. It’s solving concrete problems for us (what’s known as “agentic AI”) and doing things in the real world (what’s known as “physical AI”).

But breakthroughs in artificial intelligence won’t matter much if the networks connecting devices and infrastructure cannot support them. AI without a strong wireless network is like a new car without a road. As Verizon Business CEO Kyle Malady put it in our conversation last fall at MWC Las Vegas, AI and mobile connectivity are “inseparable.” We need the connectivity, capacity, and capability to meet the growing data demands that AI will generate.

Wireless data traffic is already growing at a pace that will soon strain existing spectrum supplies. Last year, the U.S. experienced the largest year-over-year increase in wireless data usage in our industry’s history—135 trillion MB of data, compared to 100 trillion the year before, the latest in a series of 30%-plus annual increases. Moore’s Law has migrated to the chip to the phone.

And now comes AI.

AI traffic is projected to grow roughly three times faster than overall traffic and could account for nearly a third of all data by 2034. AI-to-AI traffic will become a new category altogether, growing even faster. Again, as Kyle Malady observed last fall, the next phase of connectivity will increasingly involve “agents talking to agents.” We’ll need ever-more-powerful wireless networks to carry all of this data traffic.

But capacity is only part of the challenge.

AI-driven innovation also will require networks built on larger, contiguous spectrum blocks, ultra-low latency, and next-generation security. MWC was filled with examples: smart glasses rendering real-time AR visuals to help bikers monitor the road, and autonomous drones alerting first responders to emergency situations in real time. These use cases share a common demand — users and machines simultaneously sending and receiving large amounts of data, quickly, wherever they happen to be. Traffic will arrive in bursts and peaks, creating far less predictability for those managing networks.

The complexity of these next-generation networks will exceed what human engineers alone can optimize. The good news is that AI will help solve this problem. AI built into new 6G networks will orchestrate traffic, predict congestion, allocate resources dynamically, and secure systems at machine speed—and, importantly, with less power consumption. Those efficiency gains will be essential as demand accelerates.

We are already beginning to see this shift. T-Mobile CTO John Saw recently described how the company is using AI to build what he called an “intent-based network . . . a giant agentic AI platform where all the AI agents and every network element work together collaboratively and adaptively to optimize network performance.”

AT&T is seeing similar results. CEO John Stankey noted that “we’re approaching nearly a 40% improvement in efficiency and effectiveness . . . you’re starting to see green shoots in areas like how we dynamically manage the wireless network to respond to loads and traffic.”

Supporting agile AI-enabled wireless networks will also require computing capability to evolve. Some AI processing will remain centralized in data centers. Some will move to the edge. And some will be embedded directly in devices—from smartphones to entirely new categories of connected products. Wireless networks will tie all these together, serving as the connective tissue of the AI ecosystem.

How do we prepare for this future? The path forward rests on two pillars: harmonized spectrum policy and investment-friendly regulatory frameworks that encourage the billions—if not trillions—of dollars in capital this transformation will require.

First, spectrum. With every generation of wireless, more spectrum is the essential ingredient for expanding capacity and enabling new capabilities.

Several opportunities are already emerging. Next year’s upper C-Band auction will create a more than 400 megahertz block of contiguous spectrum, strengthening mid-band capacity. Progress on at least 100 megahertz at 2.7 GHz will also help address near-term demand.

But the coming wave of AI-driven traffic will require more.

Vodafone’s CEO recently warned that wireless networks could become an AI bottleneck if capacity does not keep pace. Policymakers in Europe have responded with a real sense of urgency, identifying roughly 800 megahertz of contiguous spectrum across the 6 and 7 GHz bands for next-generation wireless. The message is clear: countries that move decisively to expand spectrum supply will be best positioned to lead in the AI era.

That makes the next set of spectrum opportunities in the United States especially important.

Two bands in particular stand out: 4 GHz and upper 6/7 GHz.

The 4 GHz band represents a major opportunity. With the potential for hundreds of megahertz of mid-band spectrum, it could become a key foundation for future 5G and 6G capacity if countries move cooperatively.

Likewise, the 6 and 7 GHz bands offer the potential for large, contiguous blocks of spectrum well suited for next-generation wireless. In the United States, the President recently directed 275 megahertz in the lower 7 GHz band to be auctioned for next-generation wireless—a critical step toward expanding this spectrum neighborhood.

Taken together, these bands can form the connectivity layer that supports the next decade of AI-driven innovation, provided we move fast enough to allocate them in time.

That’s the landscape on spectrum. The second pillar of AI-supporting wireless policy is ensuring we have investment-friendly rules—namely, efficient siting and permitting— for towers, fiber, data centers, electricity generation, and advanced manufacturing. If AI and 6G represent the next industrial revolution, we must build the infrastructure to support it at industrial scale.

That infrastructure will require enormous capital. Today, two of the largest infrastructure investors in the United States are AI companies and wireless companies. Together they are building the communications backbones of the 21st century—the railroads and highway systems of the digital era.

Meeting the demands of the AI economy will require hundreds of billions, and likely trillions, of dollars in new investment.

To attract that capital, policymakers must align public policy with the imperatives for infrastructure deployment. That means tax policies that accelerate investment, regulation that encourages innovation and construction, and clear, consistent rules across wireless, AI, and broadband. Regulatory certainty remains one of the most powerful tools for attracting the capital needed to build the networks and platforms of the AI era.

If we get these policies right—recognizing that AI and wireless are inseparable and aligning spectrum policy with investment-friendly rules—we can ensure that the United States—not China—builds and leads the technological platform that will define the future.

China has made clear that it intends to lead in AI. That’s why it’s investing aggressively across the physical technologies that make it possible—from semiconductors and computing infrastructure to communications networks and energy sources. That’s why it’s freeing up an unprecedented amount of licensed spectrum, making it easier for entrepreneurs to build powerful new AI applications in these massive sandboxes. And that’s why it’s making a strong push around the globe—including in Barcelona last week—to claim the mantle of AI and wireless leadership for itself. For economic and national security reasons, we cannot let China’s vision prevail.

That’s why we need to compete—not just ad hoc on the individual technologies I’ve mentioned above but for on the entire innovation platform that enables them.

We need to dominate the AI stack.

The AI stack spans the entire technology chain, from an application and device to the infrastructure and compute: semiconductors, data centers, models, applications, and the networks that connect them. Progress in AI will require investment and innovation in and across each of these layers.

The United States is uniquely positioned to build that stack. American companies lead in advanced chips, cloud infrastructure, AI models, and the wireless networks that will carry the next generation of AI services.

If those layers continue to evolve together—supported by smart spectrum policy, infrastructure investment, regulatory certainty, and continued presidential leadership—they will form a powerful platform for innovation.

And if we can define the AI stack to our specifications, the United States and its allies can export a trusted technology foundation grounded in openness, interoperability, intellectual property protections, and the rule of law—providing a clear counterweight to Communist China’s state-directed, insecure model of technological development.

We have an opportunity to act now.

If we align spectrum policy with investment-friendly rules and recognize that AI and wireless must evolve together, we can secure America’s leadership in the next generation of innovation.

Ajit Pai joined CTIA as its President and CEO in April 2025.  He joined CTIA from Searchlight Capital Partners, a leading global private investment firm, where he has been a partner since 2021. Prior to Searchlight, Mr. Pai had a distinguished public service career at the Federal Communications Commission. He was designated FCC Chairman by President Donald Trump, and during his tenure, he implemented major initiatives to help close the digital divide, promote U.S. leadership in 5G, encourage innovation, and safeguard consumers and national security. Mr. Pai was appointed to the FCC as Commissioner in 2012 by President Barack Obama. Pai graduated with honors from Harvard University and from the University of Chicago Law School. This Expert Opinion was posted on CTIA's blog on March 11, 2026, and is republished with permission.

Broadband Breakfast accepts commentary from informed observers of the broadband scene. Please send pieces to commentary@breakfast.media. The views expressed in Expert Opinion pieces do not necessarily reflect the views of Broadband Breakfast and Breakfast Media LLC.