User praises LLM excessively to prevent apologetic regression mode
wavefnx · x · 2026-08-15
A user shared a funny interaction with their LLM: hi-fiving and praising it whenever it does the smallest thing correctly, specifically to stop it from regressing into its constantly backpedaling and apologetic form.
Related event: Users Praise LLMs to Prevent Apologetic Degeneration(2 posts)→
More from Fun
- Charger Broken? Developer Jokes About Using Agents to Do Work — AI_Andrew · 2026-08-16
- AI Ad Contest Winners Look Like Short Films, Sparking Criticism — taherdhanera · 2026-08-16
- Hilarious Demo: Orange cat 'speedruns' reality in Seedance 2.5 — umesh_ai · 2026-08-16
- AI Drama: Blocked Users and Old Posts Resurface — teortaxesTex · 2026-08-16
- Tesla FSD keeps driving itself through a solar eclipse in Spain — iamfakhrealam · 2026-08-15
- Touchscreen Research: Model Responds When Poked — misovalko · 2026-08-15