Wharton Professor Tests LLMs on Churchill-Style Insults

Wharton professor Ethan Mollick tested various LLMs by prompting them to generate witty, Churchill-style insults. GPT-5.6 Sol Pro outperformed others, though he noted that none of the results were truly mind-blowing.

2026-07-23 ~ 2026-07-23 · 2 related posts

• Testing LLMs with a Churchill Insult Prompt: GPT Outperforms Kimi and Gemini — emollick · 2026-07-23
• GPT-5.6 Sol Pro edges Fable in a Churchill-insult prompt test — emollick · 2026-07-23