Education Exposed the AI Trust Gap First: Lessons from GPTZero's Growth Lead
Hriday Kemburu, Growth Lead at GPTZero, says schools were the first to see a hard truth: AI can sound right and still be wrong. In 2025, GPTZero grew from nine million to twenty-four million dollars in annual recurring revenue by treating education as the front line for trust in AI-generated text.
His core point is simple. Persuasive language isn't proof of understanding. That gap is where misinformation slips in, and it showed up in classrooms before it spread across research, publishing, and the wider internet.
Education became the earliest warning system
Kemburu joined GPTZero after seeing how generative AI performs at scale in places where accuracy and citations aren't optional. As he put it: "AI has effectively watched that comic a million times, then run a million variations. It can be charismatic, confident, and wrong at scale, and now that is showing up in school, research, and publishing."
He calls GPTZero "the verification layer of the internet" and says education was ground zero. The patterns are familiar to many of us: AI slop in assignments, synthetic research, and hallucinated citations.
From classrooms to the wider internet
Early momentum came from working directly with teachers. Kemburu highlighted a "YC for AI Education" ambassador community of 10,000+ educators, which helped shape classroom AI policies and expectations for students.
Beyond schools, GPTZero partnered with HackerNoon to display "verified by GPTZero" on published posts. They now analyze 5,000+ monthly submissions and show readers AI detection disclosures so authorship and verification aren't a black box.
Verification is replacing authorship
The debate is shifting. It's no longer just "was this written by AI," but "can this be trusted." That question now drives decisions across assignments, literature reviews, and articles.
For educators, that means detection is a signal, not a verdict. The bigger move is building verification into the workflow-source checks, citation tracing, and clear rules for AI use.
What this means for educators
- Set clear AI policies: Define acceptable use, citation requirements for AI assistance, and consequences for misuse. Share examples of compliant vs. non-compliant work.
- Require source transparency: Ask students to submit prompts used, versions of outputs, and a list of tools. Make citation verification part of the rubric.
- Spot-check with multiple signals: Style shifts, odd references, impossible timelines, and broken logic matter. Use detection tools as one input, then verify citations and claims.
- Teach verification as a skill: Train students in lateral reading and source tracing. A quick primer on lateral reading pays off across subjects.
- Address hallucinations head-on: Show how confident language can still be false. Useful explainer on AI hallucinations.
- Create an escalation path: When work looks off, loop in department leads or librarians. Keep a simple paper trail of checks and outcomes to stay fair and consistent.
- Invest in staff training: Give teachers practical reps with prompts, verification, and policy. Explore AI Research Courses for educators: AI Research Courses.
Where should verification live?
Kemburu ends with an open invite: if you're seeing AI slop or trust issues, say where it shows up. That's the right prompt for schools right now.
Audit the assignments where facts and citations matter most. Tighten policies, embed verification steps, and make expectations public. The sooner students learn how to prove trust, the better their work-and your grading-gets.
Your membership also unlocks:
