GPTZero Raises 3.5Million from AI Leaders, Aims to Preserve Human Journalism

Princeton University student Edward Tian went viral building the prototype for GPTZero, the ChatGPT detection tool, out of his dorm room, which reached over 7M views on Twitter and was covered internationally in over 30 countries. Shortly after, AI researcher Alexander Cui dropped out of his PhD at the University of Toronto to join as CTO. To date, GPTZero has amassed more than 1.2 million users on the platform.

Over spring break, GPTZero raised $3.5 million from Uncork Capital and Neo, as well as Jack Altman, CEO of Lattice and brother of Sam Altman, and Emad Mostaque, CEO of Stability AI. “Our backers are betting on us to build the safeguards for large language models,” Tian said, adding that there is a growing concern that proliferating large language models are now “polluting the internet.”

Recommended: Enhancing AI: Why New Technology Must Include Diversity

In the past month, GPTZero has entered partnerships with over 40 ed-tech organizations to significantly improve AI detection accuracies on educational data, and launched the first institutional-level AI monitoring dashboard to help schools understand and craft AI policy.

This week, the startup also debuted Origin by GPTZero, a Chrome extension to assess the generation source of information in a given piece of text.

With eyes set on a journalism use-case, the team demoed Origin to Mark Thompson, former CEO of the NYT, and Tom Glocer, former CEO of Reuters, and secured investments from both newsroom executives.

“Media literacy needs a complete revolution in the AI world,” said Tian, who previously investigated election disinformation with the BBC World Service. “We’re going to lead that change with Origin.”

Latest Insights: Embrace AI to become a W.I.T.C.H. Leader

Tian and Cui have amassed an incredible team of engineers and ML researchers joining him to work on AI detection from Meta, Microsoft, and Robinhood, as well as the Vector and MILA AI research labs.

“Recent research papers have demonstrated inherent differences in AI and human writing, including AI tendencies to hallucinate false information and perpetuate biases,” said Cui. “That’s critical for end consumers to know, whether it be policy makers, financial traders, citizens, and journalists themselves.”

With Origin, the startup hopes to bolster the value of human journalism in the age of artificial intelligence by fact-checking AI-generated content, verifying citations, and exposing where else AI-generated content has appeared on the internet, to build what the team calls a ‘humanprint’ for information.

Latest Insights: Synthetic Data: A Game-Changer for Marketers or Just Another Fad?

[To share your insights with us, please write to sghosh@martechseries.com]