Reddit user tests Astra's humor with three joke examples
A Reddit user asked why LLMs remain weak at humor and shared three jokes generated by Astra Light, Medium and High tiers. The poster argued that the examples illustrate a broader limitation, but did not present them as a controlled benchmark or independently verify Astra's Codex limit consumption.
The post offers an anecdotal community test of Astra's humor generation and invites discussion about why LLM comedy may lag behind other capabilities.
I remembered that this was one of my evaluation approaches from 2 years ago to test a model's general capability. I remember wondering in the near future how bad things could get if models could begin to just spit out funny engaging content and set humanity well off on the path to Wall-E future if a cheap slop stream could be just as interesting as all of independent cultural output, and failing that, if slop could be scaled up with no limits on engagement and "perceived quality" that media companies could get people more addicted to their output than ever.
But it seemed like this was always and continues to remain well beyond reach. It's interesting to see that LLMs were able to make such a dent in an edifice so prominent as the field of mathematics and yet basically make no progress on this.
I'm trying to be cheap with my tokens... somehow utilizing Astra even from Chat is using up my codex limits. All I have to offer as examples are 3 examples. They illustrate the point well though.