Thats not a Mowhawk thats a Sharks Fin.
TIA.
OpenAI has provided its latest caption-to-image-generation model, dubbed DALL·E 2, to select users to test before it potentially opens up the technology for wider use. Named after the surrealist artist Salvador Dali and the Pixar robot character Wall-E, the model's predecessor, DALL·E, was launched last year. This software is …
To be fair, it looks like the Mohawk suggestion had just been typed in and the user hasn't yet pointy-clickied the cute li'l green arrow yet.
Doesn't look like any shark's fin I've ever seen ... More like a skateboard embedded in her skull. Or maybe an ear-flap from Floyd R. Turbo's Sunday-go-to-meetin' hat.
I'm curious as to whether DALL·E 2 can generate abstract images based on text that is *not* specifically descriptive. I played with an online service called WOMBO Dream, entering nonsense prompt text such as "Burble Burble" or descriptive but non-specific text such as "Side by side", choosing various art styles until I found something that was pleasing to me.
Because I haven't signed a licensing agreement yet I can't provide any of the images that resulted, but interested folks can always try it for themselves and see what they think.
For me, the results parallel those produced in the music composition space by products such as AIVA.