AI Detector Tool Raises Millions to Fight Online Deception

A New York-based AI detection startup, Pangram, has successfully raised $9 million in funding, signaling strong investor confidence in tools designed to combat AI-generated content. The funding round was led by Menlo Ventures, with participation from other notable investors. This capital infusion coincides with the launch of Pangram's advanced AI text detection model, Pangram 4, and a new AI image detection model.

Pangram 4 demonstrates over 99% accuracy in identifying AI-assisted writing and mixed human-AI content, also improving its ability to detect AI humanizer programs. The AI image detector, currently in research preview, is slated for a wider release soon. These developments address growing concerns about AI-generated "slop" and disinformation spreading across the internet.

Founded by Stanford AI and machine learning graduates, Pangram was established approximately two years ago in response to the surge in AI-generated content following ChatGPT's release. The company's detection system relies on a large machine learning model trained on millions of human documents, creating synthetic mirrors to learn stylistic differences and choices made by AI.

Pangram's technology is available via a $20 monthly web subscription or a Chrome extension that labels content in real-time across platforms like X, LinkedIn, and Reddit. They also offer API access, with clients including Substack, Quora, educational institutions, and recruiters. This technology aims to support human content amidst the escalating volume of AI-generated material.