How does the Pangram tool estimate if a text was generated by a machine?
Replied byChris Best
Co-founder & CEO at Substack
Niche: Technology
Revenue: Not Publicly Disclosed /month
Location: San Francisco, California, United States
Started: 2017
Pangram works by identifying specific statistical distributions and word choices that large language models naturally favor. While it can be bypassed by advanced prompt engineering, it provides a transparent signal that helps writers and readers discuss authorship.
0
From the Full Interview
This answer is part of a full interview with Chris Best, Co-founder & CEO at Substack.
Share this Answer
Found this insight valuable? Share it with your network to help others learn from Chris Best's experience.
Cite This Answer
Use this answer in your research, article, or academic work
Related Answers
How does Substack's new Pangram integration help users navigate AI content?
By Chris Best
Technology
Not Publicly Disclosed /mo
How does Baumol's cost disease apply to writing when machine text becomes free?
By Chris Best
Technology
Not Publicly Disclosed /mo
What is AI slop, and how does it define the foil for Substack?
By Chris Best
Technology
Not Publicly Disclosed /mo
How do you define the dividing line between responsible tool use and systemic gaming?
By Chris Best
Technology
Not Publicly Disclosed /mo
What is "claudefishing," and why does it feel like a betrayal to readers?
By Chris Best
Technology
Not Publicly Disclosed /mo
How does your writing process differ when you are struggling to find a concept?
By Chris Best
Technology
Not Publicly Disclosed /mo
How does the conceptual distribution of LLM ideas compare to human ideation?
By Chris Best
Technology
Not Publicly Disclosed /mo