Illinois isn’t doing so great closing its digital divide, and perhaps that’s why people in the state are picking up Starlink broadband at a faster pace than other states. According to new research from Ookla, based on Ookla’s Speedtest data, rural broadband access expanded across the U.S. in the first half of 2026, with nearly every state narrowing the gap between broadband performance in cities compared to broadband performance in rural areas. When compared to the second half of 2025, every state—with the exceptions of California, Delaware, and Illinois—made progress. New Hampshire and Virginia led the nation in gains, with New Hampshire reducing its digital divide gap from 11.5 to 4.8 percentage points.

The House passed a bill meant to protect consumers from higher electricity rates caused by data centers in one of its final votes before the midterms. The bill passed under a procedure that fast-tracks noncontroversial legislation and requires a two-thirds majority vote. The Ratepayer Protection Act passed with overwhelming bipartisan support 417-3, with three Democrats voting against it: Reps. Summer Lee (D-PA), Delia Ramirez (D-IL) and Rashida Tlaib (D-MI). Senate Majority Leader John Thune (R-SD) has signaled the Senate could take up the bill. But even if the legislation passes the Senate, it’s unlikely to immediately lower energy costs: It directs states to consider making firms that operate new data centers pay a federally set rate for high-volume electricity usage. The bill would not take full effect for two years. The vote will help vulnerable representatives argue that they have taken some action on data centers as they hit the campaign trail.

When Jessica Wachter, a finance professor at the University of Pennsylvania’s Wharton School, wanted to assess AI’s impact on the economy over the next few years, she faced a long list of business and technical uncertainties. So she started with what she calls a “remarkable fact” that is not in question: A handful of so-called hyperscalers are investing huge amounts of money to build AI data centers. Instead of trying to predict how useful and widely deployed AI models will be, she simply asked how fast the hyperscalers’ earnings will need to grow to justify their spending through 2027, when—she and her collaborator estimate—expenditures will reach nearly $1.1 trillion. It's a no-nonsense accounting approach to making sense of today’s historical AI buildout. The results are eye-opening: The AI companies will need to increase their own productivity by a factor of 2.7 to break even by 2030, accounting for the cost of capital and a 15 percent return, and depreciation of the assets. 

The Trump administration has a new villain in the battle over how fast AI development should move: effective altruism. The philosophy, which promotes the use of evidence and reason to do the most good for humanity, has helped shape and fund parts of the AI safety movement. But it has put some of its supporters at odds with President Trump's push to accelerate U.S. AI development.​ The Trump administration heavily criticized AI leaders after a number of key voices called for a slower development pace due to concerns about catastrophic risks from the technology. The Defense Department called out the EA movement by name, posting "Americanism, not effective altruism" on X. Effective altruism urges the use of data to determine how time and money can do the most good for humanity.

A mother’s viral post criticizing Meta’s AI for compiling details about her young daughters has sparked a debate on what privacy looks like in a future where technology can instantly build dossiers on people by crawling the internet. Kalie Robins, a Utah mom who posts videos about her family’s travels, uploaded a clip of herself singing in the car, her daughter in the background, when she saw a prompt underneath the post asking: “Who’s the child passenger?” Robins thought Meta wanted to verify her daughter’s identity, and she clicked to deny the request. Instead, Meta’s AI-powered chatbot provided responses about her daughter’s name, estimated age, birthdate, and interests, as well as details about another daughter who wasn’t in the video. Despite taking steps to keep her daughters’ private details offline, Meta’s AI was able to gather information from relatives’ posts about their birthdays and birth announcements.