Pangram says its new AI text detector makes only one mistake per 24,000 documents
Quick Answer
Pangram's new AI text detector, Pangram 4, boasts a 99.66% accuracy rate with only one error per 24,000 documents.
Quick Take
The model is significantly larger than its predecessor, reducing false positives by 14 times and missed AI texts by six times. The API costs $0.05 per 100 words, with additional features like image scanning included at no extra charge.
Key Points
- Pangram 4 identifies 99.66% of AI-generated text accurately.
- The model is six times larger than Pangram 3.
- False positives reduced by 14 times, missed AI texts by six times.
- API pricing is $0.05 per 100 words, with image scanning included.
- Annual revenue has grown 35 times year over year.
📖 Reader Mode
~1 min readPangram has released Pangram 4, a new model for detecting AI-generated text. The model is six times larger than its predecessor, with 14 times fewer false positives and six times fewer missed AI texts. In the company's own benchmarks, it correctly identifies 99.66 percent of all AI text and falsely flags human writing just 0.0041 percent of the time, roughly one error per 24,000 documents.
Pangram 4 can also distinguish lightly AI-polished text from fully AI-generated text, the company says. It resists "humanizer" tools that disguise AI writing as human, catching the AI component across 13 common humanizers 98.83 percent of the time.
The API costs $0.05 per 100 words, a two- to tenfold price increase depending on document length. All plans now include image scanning at no extra cost. Pangram 3 stays available until September 30, 2026. Business is booming, according to the New York Times. Annual revenue has grown 35x year over year, and monthly users jumped from 2,700 in June 2025 to 120,000 in June 2026.
— Originally published at the-decoder.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
