←back to thread

577 points simonw | 3 comments | | HN request time: 0.702s | source
Show context
croes ◴[] No.44723485[source]
I bet the training data included enough space invader cloned in JS
replies(3): >>44723515 #>>44723527 #>>44724105 #
jplrssn ◴[] No.44723515[source]
I also wouldn't be surprised if labs were starting to mix in a few pelican SVGs into their training data.
replies(3): >>44723675 #>>44723681 #>>44723878 #
1. simonw ◴[] No.44723878[source]
I'll believe they are doing that when one of the models draws me an SVG that actually looks like a pelican.
replies(1): >>44724540 #
2. __mharrison__ ◴[] No.44724540[source]
Someone needs to craft a beautifully bike donned by a pelican, throw in some seo, and see how long it takes a model to replicate it.

Simon probably wouldn't be happy about killing his multi-year evaluation metric though...

replies(1): >>44724587 #
3. simonw ◴[] No.44724587[source]
I would be delighted.

My pelican on a bicycle benchmark is a long con. The goal is to finally get a good SVG of a pelican riding a bicycle, and if I can trick AI labs into investing significant effort in cheating on my benchmark then fine, that gets me my pelican!