Karen Hao

AI Spotlight Series co-creator — Pulitzer Center Co-host, The Interface podcast — BBC Former Senior AI Editor — MIT Technology Review

Listen — profile

0:00 / 2:26

Profile

Karen Hao is the most consequential AI journalist of this era — a reporter who gets inside the companies shaping the technology and comes back with stories that force them to answer for themselves. She covered AI as the senior editor at MIT Technology Review, then moved to The Atlantic and The Wall Street Journal as a contributor. She was the first journalist to ever profile OpenAI from the inside, spending three days embedded there in 2019 and publishing a piece in 2020 that the company very much did not want.

Her reporting is slow, documentary, and unusually sourced. The 2021 investigation of Facebook’s “Responsible AI” team took nine months and produced a piece arguing that the team had been quietly redirected to work that protected growth rather than curbing the misinformation and polarization that Meta’s engagement-optimized algorithms were amplifying. At The Atlantic she wrote the definitive account of the November 2023 board firing of Sam Altman with Charlie Warzel, and she spent a year documenting the water and power footprints of the data centers behind large models — walking around a Microsoft facility in a desert town in Arizona to get the story.

In May 2025 she published Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI, based on around 300 interviews and seven years of reporting. It became an instant New York Times bestseller and won the National Book Critics Circle Award for Nonfiction. The central argument — that frontier AI labs function as a new kind of colonial empire, extracting labor and resources from the global South to feed a handful of U.S. companies — is polarizing, and Altman has publicly criticized her on social media. OpenAI declined to cooperate with the book.

For a developer learning AI, Hao is the reporter to read when you want to know what’s actually happening at the labs, not what they said in the press release. She treats AI companies the way good financial reporters treat banks: as institutions with interests, politics, and contradictions worth mapping carefully. Her work pairs naturally with researchers like Timnit Gebru, Margaret Mitchell, and Kate Crawford, who study the same systems from the inside of the field.