Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There are some quite intriguing answers to your query. We now have a thing that generates "art" in response to an input.

I suggest learning how to use a few graphics packages instead when you have a need. eg, this evening my wife wanted to print a birthday card for a dog that she is boarding. I fired up Inkscape and discovered a bug wrt handling an imported photo, scaling it and cropping it with a mask. lol! I was able to export it to a .png and print from something else.

Anyway, I was a little unfortunate and IT is still rather shakey. However, discussions about DALL-E are way more interesting than actually using it yourself. You see some seriously intelligent solutions to getting a desired image that is surely a form of programming. DALL-E does not even have a real natural language pre-processor because it is hamstrung to dump pejorative terms etc. This means that I can draw the classic, dreadful plan of the cargo hold of a slaver ship from the 17/18C and its cargo but I doubt DALL-E can.

The q&a about this thing are way more complex and interesting than its actual output! A is right, I it isn't. DALL-E is quite interesting but it really isn't intelligent. The intelligent bit is getting the input right for the desired output. Perhaps another model could be developed for that.



Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: