Users praise models such as Claude for brief responses and straightforward reasoning on fast tasks, but others criticize the precision and highlight dangers like potential codebase damage.
Based on 8 visible X reactions from 76 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@goodside I like that it just says "Done." and doesn't share any of it's thoughts. reminds me of my husband. AGI.
@goodside fair enough. refusing to roleplay an hour of thinking is more honest than most progress bars.
@SCHIZO_FREQ obviously trained on lazy time constrained monkeys
@SCHIZO_FREQ Except the 10% of cases #2 will completely mess up your code base
@goodside I like that it just says "Done." and doesn't share any of it's thoughts. reminds me of my husband. AGI.
@goodside fair enough. refusing to roleplay an hour of thinking is more honest than most progress bars.
@SCHIZO_FREQ obviously trained on lazy time constrained monkeys
@goodside Common Claude W
@SCHIZO_FREQ the accuracy 💀
LLMs seem to have poor intuition for how much time their thinking requires. ChatGPT 5.6 Pro: https://x.com/goodside/status/2076779839638347933/photo/1
Users praise models such as Claude for brief responses and straightforward reasoning on fast tasks, but others criticize the precision and highlight dangers like potential codebase damage.
Based on 8 visible X reactions from 76 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@SCHIZO_FREQ the accuracy 💀