Subscribe for premier reporting on free speech, privacy, Big Tech, media gatekeepers, and individual liberty online.

US government plans to develop AI that can unmask anonymous writers

If you’re tired of censorship and surveillance, subscribe to Reclaim The Net.

The Office of the Director of National Intelligence (ODNI) said that the Intelligence Advanced Projects Activity (IARPA) is working on a program to unmask anonymous writers by using AI to analyze their writing style which is seen as potentially being as unique as a fingerprint.

“Humans and machines produce vast amounts of text content every day. Text contains linguistic features that can reveal author identity,” IARPA said.

If successful, IARPA believes the Human Interpretable Attribution of Text Using Underlying Structure (HIATUS) program could identify a writer’s style from different samples and modify those patterns to further anonymize the writing.

“We have a strong chance of meeting our goals, delivering much-needed capabilities to the Intelligence Community, and substantially expanding our understanding of variation in human language using the latest advances in computational linguistics and deep learning,” said HIATUS program manager Dr. Timothy McKinnon.

IARPA said that it will also develop explainability standards for the program’s AIs.

ODNI said HIATUS could have multiple applications, including combating foreign influence activities, protecting writers whose work may put a target on their backs, and identifying risks to counterintelligence. And according to McKinnon, the program can identify if text was generated by a machine or written by a human being.

But it is not IARPA’s work to turn HIATUS into something usable. The agency’s work is only to develop the technology.

If you’re tired of censorship and surveillance, subscribe to Reclaim The Net.

Read more

Join the pushback against online censorship, cancel culture, and surveillance.

Already a member? Login.