Pangram Labs, a Brooklyn-based AI content detection platform, has raised $9M in funding led by Menlo Ventures with participation from Haystack, ScOp Venture Capital, Script Capital, and Cadenza. The company provides AI-generated content detection at 99.98% accuracy with a false positive rate of 1 in 10,000, using proprietary deep learning models trained on millions of human and AI documents. It supports detection across major LLMs, 20+ languages, AI-assisted writing, plagiarism checking, and image detection in research preview. The capital will accelerate product development including Pangram 4 and multimodal expansion.
AI Detection Demand Rises With Backlash
The timing aligns with growing skepticism toward existing tools. GPTZero raised $13.5M in seed and Series A funding by June 2024 before its acquisition by Superhuman in June 2026. Turnitin faces institutional pushback over false positives, with some universities disabling its AI detection. Pangram's approach prioritizes minimal false accusations through granular detection of fully AI-generated, AI-assisted, and humanized text.
False Positives Undermine Trust in Detection
Universities and publishers require reliable tools because erroneous accusations carry legal and reputational risks. Independent studies show Pangram achieving near-zero false positive rates on human documents while competitors drop to 0% accuracy on advanced models like GPT-4o Deep Research. One analysis of 1,163 master's theses flagged 45.5% for likely AI content, highlighting the scale of undisclosed use.
Proprietary Models Outperform on Accuracy
Pangram trains on synthetic mirrors of human documents to isolate stylistic differences, enabling detection even after humanizer tools attempt obfuscation. It also distinguishes specific LLMs and classifies AI assistance levels. The platform integrates with LMS systems like Canvas and Google Classroom plus browser extensions for real-time scanning on platforms including Substack.
"With LLMs, the cost of producing language collapses to zero, and the cost of trusting language goes to infinity. Pangram has emerged a clear #1 in quality."
Menlo Ventures Backs Technical Leadership
The round follows earlier seed funding, bringing total capital to roughly $13M. Menlo Ventures partner Deedy Das cited the company's position across Substack and Quora as validation. The investor profile signals conviction in infrastructure for content authenticity rather than pure growth capital.
Market Expands to Multi-Billion Scale
The AI detector market stands at $749.8M in 2026 and is projected to reach $5.2B by 2033 at a 32% CAGR. AI Content Detection Software is expected to grow from $2.20B to $8.56B over the same period. Competitors include Originality.ai, which raised about $1M, and Copyleaks with $7.1M total. Regulatory pressure for AI transparency and exploding generative content volumes are driving institutional and enterprise adoption.
Partnerships Signal Platform Momentum
Recent integrations with NewsGuard for identifying AI content farms and Substack for on-platform scanning demonstrate traction beyond education. The company has also open-sourced models and published peer-reviewed research at ICLR 2026, distinguishing it from purely commercial rivals.