•
Jun 10
Fiber Connect is the premier fiber optic tradeshow, providing a platform to stay up-to-date on developments in the broadband industry. The 2025 event yielded several interesting takeaways, with artificial intelligence (AI) and digital twins capturing a number of attendees’ and speakers’ attention in Nashville. Here are four of our top observations from Fiber Connect 2025.
The Broadband Equity, Access and Deployment (BEAD) program is a $42.45 billion federal grant program aiming at expanding high-speed internet access across the United States. It focuses on funding planning, infrastructure deployment and adoption programs to ensure that unserved and underserved areas are connected to high-capacity, future-ready broadband infrastructure.
The new administration has been working to cut government red tape that has previously slowed infrastructure construction. As a result, the funding itself is undergoing some reforms. On June 6, the new administration made new rules for the program, requiring all participants to modify their bids within the next 90 days.
Yet extending fiber into rural communities remains a high priority for a number of parties. The reasons are numerous: fiber promotes higher business growth, can drive increased self-employment, and can subsequently deliver higher GDP and per-capita income.
Source: here
Several state broadband chiefs shared both their frustration and trepidation regarding anticipated changes to the multibillion-dollar program. Despite delays in BEAD funding, privately funded fiber deployments — backed by private equity and corporate capital — are accelerating.
Middle-mile networks connect “last mile” (i.e., local) networks to other service providers, major telecommunications carriers and the greater internet. Although they do not typically connect the majority of end users, middle-mile ownership is critical because it enhances the bond between internet service providers (ISPs) and the community.
With middle-mile support, an ISP can connect anchor institutions such as schools, hospitals and libraries. In some cases, their middle-mile assets can even enable open-access partnerships and broadband access programs, thereby reinforcing their role as local providers and community partners.
In many areas throughout the U.S., there is no middle-mile network close enough to connect to the last-mile infrastructure. Alternatively, the only nearby middle mile is privately owned by a large ISP that can selectively determine the business criteria for serving the area.
Source: here
Additionally, middle-mile networks are not adequately scaled to support modern broadband demands. This challenge is further intensified by the rapid expansion of last-mile operators and ISPs. Middle-mile networks serve outdated assumptions about household bandwidth needs. “While public reports suggest peak-hour usage of 15–20 Mbps, real-world observations show active households often exceeding 100 Mbps,” wrote the FBA.
Source here.
To effectively meet current broadband needs, operators must redesign their networks to reflect actual usage patterns. With billions in federal broadband infrastructure funding becoming available, the report urges decision-makers to prioritize middle-mile development to close the digital divide once and for all. There is an urgent need to scale and redesign middle-mile networks to meet the increasing demand for fiber across the U.S., a priority that AI solutions can help telecom and municipal leaders achieve.
Fiber is essential in our always-on world, particularly in light of its role in supporting data center buildouts and forming the backbone of AI infrastructure. As a result, there is a growing need to facilitate dialogue between the fiber and data center communities, including where emerging quantum-computing discussions are blossoming. This opens a strategic opportunity to apply AI-driven economic modeling in broadband policy decisions.
Currently, many fiber companies lack a clear strategy for developing an ROI-focused business case regarding where and how to deploy their infrastructure. By leveraging public and private data sources, crowdsourced data (e.g., Ookla, Meta mobility datasets and others), and tracking data center developments, AI can model and predict optimal locations for last-mile infrastructure. This development creates a strategic opportunity to apply AI-driven network and infrastructure planning methodologies.
In part because of ongoing BEAD discussions, one of the hottest topics and biggest challenges is how to provide sustainable connectivity for populations that cannot afford traditional broadband. In rural areas like Bandera County, Texas, local cooperatives are concerned that communities may disappear entirely without access to digital services. Similar concerns are emerging in remote parts of Hawaii, where a lack of connectivity has left children without home internet and families dependent on couch-surfing.
Connectivity offers numerous benefits to communities, including the integration of telehealth services, online learning options and government-backed digital inclusion programs. These are just a few examples of connected services that promote the long-term viability of a community.
As fiber planning continues, providers and their ecosystem partners must consider these digital services in their models to ensure that their infrastructure is future-ready and supports ongoing efforts to bridge the digital divide.
AI and digital twins for telecom network design and optimization are proven tools to support network design, capacity planning and quality of service (QoS) and quality of experience (QoE) modeling. AI can also automate site surveys, generate network designs and produce engineering-grade drawings (e.g., CAD files), dramatically accelerating time-to-deploy.
With fiber companies looking to accelerate and improve their decision-making processes, AI and digital twins can be critical enablers to optimize rollout plans based on business value and usher-in the industry’s next phase of growth.
Other articles that may interest you
