i've been testing something that works so well im not that excited to use codex
@bentossell
-
ChatGPT Audio Transcription Bug Report
By
–
weird bug on chatgpt I was using audio and the transcription ended up as this…
-
LLM with Dashboard Tool Access for One-Click Decisions
By
–
not 100% but maybe if I fed that list of things into an imaginary llm and gave it tool access to the dashboard, all the knobs and switches, I one click accept I get it’s hard but I don’t want to have to click off to read docs to then try remember where I am, come back etc.
-
AI Onboarding Should Do the Work For You, Not Tell You
By
–
how many have you gone through?
from going through (any) did you ever get lost in the product again?
onboarding just tells you what YOU should do, but AI should do it FOR you -
AI Will Be Proactive, Not Reactive – Guided User Experiences
By
–
ai will be proactive not reactive customers having to learn your product should disappear you’ll be guided, prompted to click buttons, etc
-
o3 Model Performance Varies With Different Instructions
By
–
Yeh I dunno if it’s some memory or instruction that veers it off. I felt like past instructions helped the models but with o3 it performs differently when changing instructions
-
o3 Model Experiencing High Hallucination Rates
By
–
o3 is hallucinating a lot, not sure if recent or always been true thought it was the smartest model?
-
Tools Should Implement Security Fixes, Not Users
By
–
its for the tools to implement, not the users e.g. supabase security advisor – one-click fix for security issues