The founder who built the internet's most popular AI detector just sold to an AI-first company and says if he could do college over, he'd skip computer science entirely.
The Summary
- Edward Tian built GPTZero over winter break in his senior year at Princeton, turned it into an overnight success, then sold it three years later to Superhuman, a decidedly pro-AI company
- He credits "artificial scarcity" as a key growth driver, not just the product itself
- Despite building an AI detection empire, Tian now says he'd major in humanities over computer science if starting college today
- The sale to Superhuman caught people off guard: an AI detection company joining forces with AI maximalists
The Signal
Edward Tian was lined up for a visual investigations fellowship and ready to start a journalism career when ChatGPT dropped. He and his high school friend Alex, then pursuing a Ph.D. in computer science at the University of Toronto, built GPTZero during winter break. They launched it online. The growth was immediate.
Three years later, they sold to Superhuman. Not a defensive AI company. Not an academic consortium. A productivity tool known for being aggressively pro-AI. The move surprised people who thought GPTZero existed to keep AI off the internet.
"We didn't build GPTZero to say there should be no AI on the internet. I think that's the biggest misunderstanding about why we started our company."
But the arc from detection tool to acquisition reveals something more interesting than a simple exit. Tian attributes part of GPTZero's success to creating "artificial scarcity" around the product. In a market flooded with free AI tools, limiting access created demand. It's a playbook that runs counter to the "release fast, scale faster" doctrine of most AI startups. Scarcity works when everyone else is giving things away.
The real tell is what Tian says about his own education. He's 26 and already thinks he picked the wrong major. Computer science got him the technical chops to build a company. But now, watching AI reshape writing and creative work, he'd choose humanities. He sees the future tilting away from code literacy and toward something else: understanding what makes human expression valuable when machines can mimic it perfectly.
Key tensions:
- Built an AI detector, sold to an AI accelerator
- Became successful through scarcity in an abundance economy
- Trained as an engineer, now advocates for humanities
That's not a founder having regrets. That's a founder who built in one paradigm and sees the next one forming. The skills that helped him capitalize on the first wave of AI panic won't be the skills that matter when AI writing is ubiquitous and the question shifts from "is this AI?" to "does this matter?"
The Implication
If you're betting your career on detecting AI or fighting it, you're already behind. The Superhuman acquisition signals that even the detection layer gets absorbed into the AI stack. The real edge is going to be understanding context, taste, and why certain human outputs still command attention when the bots can produce infinite variations.
Tian's pivot from journalism to detection to humanities advocate isn't random. It's pattern recognition. Watch what the people building AI tools actually do with their time and money. If they're moving toward liberal arts, curation, and storytelling, that's where the scarcity is heading next.