Users love GPT-5.6 generating a working retro word game called Don't Discuss Goblins from a vague prompt because such toy projects reveal clever mechanics and real AI capabilities better than benchmarks.
Based on 3 visible X reactions from 7 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@emollick The fact that 'dumbest project' still produces a working game with a clever mechanic is the actual signal. Nobody benchmarks that, but it's the thing that changes what a solo dev can ship in a weekend.
@emollick Dumb toy projects are secretly the best eval. A clever mechanic from a vague prompt shows taste, and taste is the thing benchmarks still can't measure.
@emollick Name is hilarious. I love dumb ai project like this they actually show what it can do
Now that GPT-5.6 is live I can share demos, so I am starting with the dumbest project I had it do during the preview: make a game called "Don't Discuss Goblins." From a prompt, it generated ideas and came up with a pretty clever mechanic and graphics: https://dont-discuss-goblins.netlify.app/game/ https://x.com/emollick/status/2075269373426536691/video/1
Users love GPT-5.6 generating a working retro word game called Don't Discuss Goblins from a vague prompt because such toy projects reveal clever mechanics and real AI capabilities better than benchmarks.
Based on 3 visible X reactions from 7 accounts; directional sample.
Ask a question below.
Published answers will appear here.