Can You Detect AI?

Substack has launched Pangram, an AI detector to combat “AI slop,” but its effectiveness is questionable as AI can be prompted to bypass it. The article outlines common AI writing tells and provides methods to write more naturally, even when using AI tools. It emphasizes that human intuition and specific stylistic markers are more reliable than current AI detection technology.

• Substack is using Pangram, a leading AI detector, to combat AI-generated content.
• Pangram is effective at detecting AI, but can be bypassed with specific prompts and minor text edits.
• Common AI writing tells include uniform sentence length, repetitive structures, and overused “AI words.”
• To avoid sounding like AI, humans should use contractions, varied sentence structures, concrete imagery, and natural language.
• The article suggests that human-written text often has a “burstiness” and natural flaws that AI detectors miss.
• The author shares a method for using AI to write newsletters, including specific prompts and skills to bypass detection.
  https://bender.layer3.press/articles/c1cc954c-80ec-4cff-87b6-8430cdd0b45b