Pangram Raises $9M to Detect AI-Generated Content as Internet Floods with AI Slop

New York-based startup Pangram has raised $9 million in a funding round led by Menlo Ventures to expand its AI content detection technology. The company simultaneously launched its next-generation AI text detection model, Pangram 4, which it claims is over 99% accurate at identifying AI-assisted writing and mixed human-AI content, and an AI image detection model, Pangram Image, currently in research preview. Co-founders Max Spero and Bradley Emi, Stanford AI and machine learning graduates, launched Pangram about two years ago following ChatGPT's release. The detection system is trained on tens of millions of human documents and uses a synthetic mirror technique to learn stylistic differences between human and AI writing. Pangram offers a $20-per-month subscription, a Chrome extension that labels content on social media platforms in real time, and an API. Notable customers include Substack and Quora. In testing, the text detector performed well but occasionally mislabeled human-written sentences as AI. The image detector can identify AI-generated images across different models and even detect AI images within real photos. Spero emphasized that the goal is not to witch-hunt AI users but to preserve human content against the growing tide of AI-generated material.