House Subcommittee Witnesses Disagree on AI for Broadband Maps

WASHINGTON, November 14, 2023 – Experts disagreed on the potential for artificial intelligence to aid broadband mapping efforts at a House hearing on Tuesday.

Courtney Lang, a vice president at tech industry trade group ITI, said AI could be used to improve the quality of current broadband maps.

A machine learning model could do that by using past data to identify buildings that are likely to be accurately marked as having adequate broadband, according to Lang.

“It’s a really interesting use case,” she said.

Broadband mapping is a difficult task. The Federal Communications Commission’s broadband map is on its third version, undergoing revisions as consumers submit challenges to provider-reported broadband coverage data. The Biden administration’s $42.5 billion broadband expansion program requires states to administer a similar ground-truthing process before allocating any of that cash.

But Nicol Turner Lee, director of the Brookings Institution’s Center for Technology Innovation, urged caution.

“We have to be careful that we might not have enough data,” she said.

In rural parts of the country, data can be sparse and low-quality. Both factors would make machine learning ill-suited to the task of flagging potential inaccuracies, according to Lee.

She urged lawmakers to exercise restraint when using AI for “critical government functions,” like the broadband maps used to determine where federal grant money will go.

The witnesses spoke at a House Communications and Technology Subcommittee hearing on using AI to enhance American communication networks.