Pangram Raises $9M to Combat AI-Generated Content Flooding the Internet with Advanced Detection

Pangram Secures $9M to Expand AI Detection Capabilities

New York-based AI detection startup Pangram raised $9 million in a funding round led by Menlo Ventures, with participation from Haystack, ScOp Venture Capital, Script Capital, and Cadenza1

1

. The investment brings the company's total funding to nearly $13 million after raising $2.7 million in June 20254

4

. Founded by Stanford AI graduates Max Spero and Bradley Emi approximately two years ago, Pangram emerged after ChatGPT's launch opened the floodgates for AI-generated content across the internet1

1

.

Pangram 4 Delivers Enhanced Accuracy Against AI Slop

Coinciding with the fundraise, Pangram launched its next-generation AI text detection model, Pangram 4, which the company claims achieves over 99% accuracy at identifying AI-assisted writing and mixed human-AI content1

1

. The new model demonstrates major improvements in detecting AI-assisted writing and shows greater robustness against AI humanizer programs that attempt to disguise machine-generated text as human writing3

3

. According to internal benchmarks, Pangram 4 maintains an industry-leading false positive rate of 0.0041%, or approximately once for every 24,000 documents4

4

. Spero, who calls himself the internet's "slop janitor," told The New York Times that the company's AI text-detection model can accurately identify 9,999 out of 10,000 times whether text was AI-generated2

2

.

How Pangram's Classifier Model Works

Pangram's AI detection system functions as a large machine learning model trained on tens of millions of known human documents1

1

. The company created synthetic mirrors for each document, replicating the topic, length, and tone of voice but written by frontier LLMs1

1

. This classifier model learns the stylistic differences and choices that AI makes consistently, enabling it to identify AI-generated content with high confidence without relying on copy-paste metadata or hidden watermarks5

5

. Menlo Ventures partner Deedy Das explained the problem: "With LLMs, the cost of producing language collapses to zero, and the cost of trusting language goes to infinity"3

3

.

AI Image Detection Model Enters Research Preview

Pangram also unveiled an AI image detection model, currently available via research preview, which the company plans to release more widely in coming weeks1

1

. The AI image detection model achieves 99.5% accuracy in internal benchmarks and can identify images created by major generative AI platforms including OpenAI's GPT Image, Google's Gemini Nano Banana, Midjourney, FLUX, and Grok Imagine4

4

. Bradley Emi, Pangram's CTO, noted that AI image detection presents unique challenges because real and AI content is often blended, such as in photos of AI-generated ads on billboards3

3

.

Growing Demand Amid Rising AI Content Concerns

The funding comes as AI content floods the internet, with consequences ranging from embarrassment to serious legal sanctions. A Canadian politician recently read an AI prompt aloud during a speech to lawmakers, while lawyers have faced sanctions and fines for using fake citations created by ChatGPT1

1

. Institutional pushback is mounting: the open-access archive arXiv introduced a new enforcement policy this year stating that submissions containing evidence of unreviewed LLM output can trigger a one-year submission ban1

1

. Research from Let's Enhance shows that human detection rates for AI images hover around 63.7%, dropping to near 29% for high-capability generators like FLUX4

4

.

Subscription Access and API Integration

Users can access Pangram via a $20-per-month subscription on the web or through a Chrome extension that automatically labels posts in real time on platforms including X, LinkedIn, Substack, Reddit, and Medium1

1

. The extension provides a feed health score with a percentage breakdown of human versus AI content on screen1

1

. Pangram also offers its technology via API, with Substack recently integrating Pangram's technology to show readers which authors write newsletters using AI1

1

. Other API customers include Quora, schools and universities, publishers, agents, and recruiters1

1

.

Competitive Landscape and Reliability Questions

Pangram faces competition from other AI detection tools including Winston AI, Originality.ai, Copyleaks, and GPTZero, each building its own detector to address growing demand1

1

. However, broader concerns about AI detector reliability persist. MIT Sloan Teaching and Learning Technologies points out that the technology remains "far from foolproof" with high error rates that can lead to false accusations4

4

. The Mozilla Foundation found that detector tools can be biased against non-native English speakers4

4

. Despite these industry-wide challenges, Pangram positions itself as having emerged as "a clear #1 in quality" and is trusted across Substack, Quora, and other platforms, according to Menlo Ventures3

3

.