New York-based startup Pangram has secured $9 million in funding, led by Menlo Ventures, to advance its AI detection technology aimed at identifying AI-generated content flooding the internet. Founded by Stanford AI graduates Max Spero and Bradley Emi roughly two years ago, the company recently launched Pangram 4, an AI text detection model claiming over 99% accuracy in spotting AI-assisted and mixed human-AI writing, along with a research preview of an AI image detection model. Unlike watermark or metadata-based systems, Pangram’s approach uses a machine learning model trained on millions of human documents paired with synthetic AI-generated counterparts to detect subtle stylistic differences inherent in AI writing.

Pangram focuses not only on fully AI-generated texts but also on content with varying degrees of AI assistance, such as human-authored drafts polished by AI. Spero highlights the importance of transparency and disclosure in AI use, asserting that readers benefit from knowing whether content was produced or edited by AI tools. The startup offers a subscription service and a Chrome extension that labels AI-generated posts in real time across social platforms like X, LinkedIn, Reddit, and Medium. Additionally, Pangram's technology is integrated via API by publishers such as Substack, enabling readers to see which newsletters are AI-assisted.

The growing prominence of AI-generated content has led to real-world consequences, from embarrassment over public speeches incorporating AI prompts to legal sanctions against lawyers using AI-fabricated citations. Pangram enters a competitive field that includes companies like Winston AI and GPTZero, offering tools designed to challenge the increasing volume of AI slop overriding genuine human content online and in academic and legal environments. While not perfect, Pangram’s models have demonstrated strong performance in tests, including detecting AI image manipulations embedded in real photos and identifying AI-text even after human edits.

Looking ahead, Pangram aims to expand access to its AI image detection capabilities and fuel efforts to maintain the integrity and trustworthiness of information online. Spero warns that unchecked AI content proliferation risks drowning out authentic human voices, emphasizing the need for effective discrimination tools to preserve quality and reliability in digital communication. The startup’s $9 million raise will help scale these detection technologies as demand for AI content identification accelerates across multiple industries.