Jeff Kao

Computational Journalist

Photo of Jeff Kao

Jeff Kao is a computational journalist at ProPublica who uses data science to cover technology and artificial intelligence. He used natural language processing techniques to uncover 1.3 million fake comments submitted to the FCC in its proceeding repealing net neutrality. This work was cited in the Washington Post, Fortune Magazine and engadget, among other publications, and by members of the U.S. Senate. He has appeared as a data scientist in the New York Times and on the WNYC program Science Friday.

Kao previously worked as a machine learning engineer at Atrium LTS, where he developed natural language processing systems for legal services. He holds a law degree from Columbia Law School, where he was the editor-in-chief of the Columbia Science and Technology Law Review, and a bachelor’s degree in systems design engineering from the University of Waterloo.

The Disinfomercial: How Larry King Got Duped Into Starring in Chinese Propaganda

The broadcasting icon’s fake interview with a Russian journalist went viral on social media, spread by accounts tied to China’s government.

There’s Been a Spike in People Dying at Home in Several Cities. That Suggests Coronavirus Deaths Are Higher Than Reported.

Coronavirus death counts are based on positive tests and driven by hospital deaths. But data from major metropolitan areas shows a spike in at-home deaths, prompting one expert to say current numbers were just “the tip of the iceberg.”

How China Built a Twitter Propaganda Machine Then Let It Loose on Coronavirus

ProPublica analyzed thousands of fake and hijacked Twitter accounts to understand how covert Chinese propaganda spreads around the globe.

Millions of Americans’ Medical Images and Data Are Available on the Internet. Anyone Can Take a Peek.

Hundreds of computer servers worldwide that store patient X-rays and MRIs are so insecure that anyone with a web browser or a few lines of computer code can view patient records. One expert warned about it for years.

Aggression Detectors: The Unproven, Invasive Surveillance Technology Schools Are Using to Monitor Students

In response to mass shootings, some schools and hospitals are installing microphones equipped with algorithms. The devices purport to identify stress and anger before violence erupts. Our testing found them less than reliable.

Methodology: How We Tested an Aggression Detection Algorithm

An in-depth look at software that claims to spot aggression from your voice.

The Trade Secret: Firms That Promised High-Tech Ransomware Solutions Almost Always Just Pay the Hackers

As ransomware attacks crippled businesses and law enforcement agencies, two U.S. data recovery firms claimed to offer an ethical way out. Instead, they typically paid the ransom and charged victims extra.

We’re Reporting on Ransomware. Do You Know Something About an Attack?

Has your organization been hit by ransomware? Did you hire a data recovery firm? Do you know how an attack works from the inside? We’d like to hear from you.

Georgia Officials Quietly Patched Security Holes They Said Didn’t Exist

A ProPublica analysis found that the state was busily fixing problems in its voter registration hours after the office of Secretary of State Brian Kemp, the Republican candidate for governor, had insisted the system was secure.

File-Sharing Software on State Election Servers Could Expose Them to Intruders

A ProPublica analysis found election computer servers in Wisconsin and Kentucky could be susceptible to hacking. Wisconsin shut down its service in response to our inquiries.

Follow ProPublica

Latest Stories from ProPublica

Her Stepfather Admitted to Sexually Abusing Her. That Wasn’t Enough to Keep Her Safe.

Oil Companies Are Profiting From Illegal Spills. And California Lets Them.

Electionland 2020: USPS Mailers, Pandemic Voting, Get Out the Vote Efforts and More

Poorly Protected Postal Workers Are Catching COVID-19 by the Thousands. It’s One More Threat to Voting by Mail.