- Despite the uncertainty around BEAD, Zayo’s Bill Long expects state middle-mile efforts will keep ramping up
- Technologies like satellite won’t solve the “bandwidth doom-loop” that’s affecting mid-sized cities, he said
- He also thinks 2025 will be ‘transformative’ in dealing with permitting constraints
We’re all waiting with bated breath to find out the fate of fiber and the BEAD program in 2025. But states will keep going full steam ahead with their middle-mile fiber ambitions, said Bill Long, chief product officer at Zayo.
National Telecommunications and Information Administration (NTIA) has already awarded nearly $1 billion in grants from its Middle Mile Program. Some of the recipients are states actively building out their own middle mile networks, like California, Kansas, Maine and Nevada.
In the coming year, Long anticipates we’ll see more states jump on the bandwagon and start middle-mile broadband initiatives, using not just federal dollars but also their own resources.
Folks in the industry are concerned there isn’t enough money for middle mile to go around, and it can also be costly for last-mile providers to access middle mile infrastructure. Still, Long is optimistic states can balance the funding they need between middle-mile and last-mile deployments.
“The concentration of [BEAD] funds in red states versus blue states also bodes well for continued middle mile [funding],” he said.
A key question the industry’s asking is whether BEAD under Trump 2.0 will do away with the fiber preference in favor of other technologies that are cheaper and faster to deploy, like satellite.
While satellite is an option to connect “very rural locations,” it won’t help address the “bandwidth doom loop” that’s affecting mid-sized cities, said Long.
The “doom loop” refers to when towns and cities don’t have the bandwidth capacity to attract industry. ISPs meanwhile don’t want to build networks where there aren’t any businesses.
These medium-sized cities, which typically have a population between 20,000-40,000, need to connect distribution centers that belong to the likes of Amazon, FedEx and Walmart.
Those companies are “spending hundreds of millions of dollars to automate those distribution centers with lots of robots, automation and AI,” Long said. “If they don’t have high-bandwidth connections to those locations, they don’t work.”
He anticipates it will take a year or two to build middle-mile networks at scale. But it’ll probably take 5-10 years before “the dynamics of the local economies will really start to change,” he told Fierce.
“There’s going to need to be a lot more money to bring people not just into the digital economy but to the AI-enabled economy,” Long added.
More power for AI, less for permitting
2024 saw increased activity between telcos and cloud providers to get ready for AI. Lumen won a big contract with Microsoft to supply private fiber connectivity between the latter’s data centers, and Windstream recently told Fierce it’s seeing more demand for higher dark fiber counts to support GPU capacity.
Safe to say this activity will require lots of power. According to Long, the cost of power transmission is “four to 10 times” higher than the cost to deploy fiber.
“I think you’re going to see a lot of demand in bringing fiber to locations where there’s big units of power,” he said.
And as AvidThink’s Roy Chua previously told Fierce, companies that have dark fiber capacity on hand can build data centers in locations where power is cheaper.
We’ve heard plenty of complaints from operators on how cumbersome the permitting process is. But the issue doesn’t solely lie with fiber deployment. Long noted permitting constraints exist across the entire “AI infrastructure ecosystem,” including data centers and power transmission.
Shortly after the election, OpenAI published a “blueprint” for U.S. infrastructure, which, among other things, suggested the creation of “AI Economic Zones.” According to OpenAI, states in these zones can incentivize faster permitting and approvals for AI infrastructure like data centers, solar arrays and wind farms.
Regulatory change could be afoot on the permitting front. Future FCC Chairman Brendan Carr has been pretty vocal about cutting down on government regulations “that are overly cumbersome or outdated.”
As a result, “’25 is going to be a pretty transformative year” for solving permitting problems, Long concluded.
