Pangram, a startup focused on detecting AI-generated content, has raised $9 million as the internet grapples with an unprecedented flood of synthetic text.
The company’s technology analyzes text at the statistical and structural level to determine whether a human or an AI model wrote it. As generative AI tools churn out everything from blog posts to product reviews at scale, publishers and platforms are racing to deploy detection systems that can separate authentic content from machine-generated material.
Pangram’s approach differs from earlier detection tools that relied on easily bypassed pattern matching. The company claims its models maintain high accuracy even against the latest generation of AI writing models, which have become significantly harder to distinguish from human authors.
The funding comes as regulators in the US and EU increasingly scrutinize AI-generated content, particularly in contexts where authenticity matters for consumer protection, election integrity, and intellectual property. Several publishing platforms and social media companies have already begun trialing Pangram’s technology, according to the company.
The $9 million seed round will allow Pangram to expand its research team and build out commercial operations ahead of what the company expects will be a wave of regulatory mandates around AI content labeling.
