
If you spend any time at all on social media, you’ll have glimpsed the work of DALL‑E, OpeÂnAI’s now-famous artiÂfiÂcial-intelÂliÂgence engine that genÂerÂate images from simÂple text descripÂtions. A velociÂrapÂtor dressed like Travis BickÂle, AmerÂiÂcan GothÂic starÂring Homer and Marge SimpÂson, that astroÂnaut ridÂing a horse on the moon: like any the-future-is-now moment, espeÂcialÂly in recent years on the interÂnet, DALL-E’s rise has proÂduced a host of artiÂfacts as impresÂsive as they are ridicuÂlous. Now you can try to top them in both of those dimenÂsions yourÂself, since not just DALL‑E but the new, improved, highÂer-resÂoÂluÂtion DALL‑E 2 has just opened for pubÂlic use.

“How do you use DALL‑E 2?” You might well ask, and CreÂative Bloq has a guide for you. “The tool genÂerÂates art based on text prompts,” it explains. “On the face of it, that couldÂn’t be more simÂple. Once you’ve comÂpletÂed the DALL‑E 2 sign up to open an account, you use the proÂgram in your browsÂer on the DALL‑E 2 webÂsite. You type in a descripÂtion of what you want, and DALL‑E will creÂate the image.”
Of course, some prompts proÂduce more visuÂalÂly interÂestÂing results than othÂers. The guide recÂomÂmends that you conÂsult the DALL‑E 2 prompt book, which gets into how best to phrase your descripÂtions in order to inspire the richÂest comÂbiÂnaÂtions of subÂject, texÂture, style, and form.

“Even the creÂators of DALL‑E 2 don’t know what the tool knows and doesÂn’t know. Instead, users have to work out what it’s capaÂble of doing and how to get it to do what they want.” And indeed, that’s the part of the fun. DALL-E’s own interÂface recÂomÂmends that you “start with a detailed descripÂtion,” and with a litÂtle experÂiÂmenÂtaÂtion you’ll disÂcovÂer that speciÂficiÂty is key. The renÂderÂings of “an eight-bit NinÂtenÂdo game designed by Hiroshige” and “a cyberÂpunk downÂtown Los AngeÂles scene paintÂed by RemÂbrandt” strike me as credÂiÂble enough for a first effort, but adding just a few more words opens up entireÂly new realms of surÂprise and inconÂgruity.

Just above, we have two of DALL-E’s infiÂniteÂly many posÂsiÂble attempts to visuÂalÂize “the covÂer of an old Ernest HemÂingÂway pulp novÂel about the advenÂtures of David Bowie.” Though the designs look entireÂly plauÂsiÂble, the titles highÂlight the techÂnolÂoÂgy’s already-notoÂriÂous inabilÂiÂty to come up with intelÂliÂgiÂble text. OthÂer limÂiÂtaÂtions of the newÂly pubÂlic DALL‑E, accordÂing to Ars TechÂniÂca’s Benj Edwards, include the requireÂment to proÂvide your phone numÂber and othÂer inforÂmaÂtion in order to sign up, the ownÂerÂship of the genÂerÂatÂed images by OpeÂnAI, and the necesÂsiÂty to purÂchase “credÂits” to genÂerÂate more images after you’ve run through your iniÂtial free 50. Still, there’s nothÂing quite like typÂing in a few words and sumÂmonÂing up works of art no one has ever seen before to make you feel like you’re livÂing in the twenÂty-first cenÂtuÂry. You can sign up here.
RelatÂed conÂtent:
Based in Seoul, ColÂin MarÂshall writes and broadÂcasts on cities, lanÂguage, and culÂture. His projects include the SubÂstack newsletÂter Books on Cities, the book The StateÂless City: a Walk through 21st-CenÂtuÂry Los AngeÂles and the video series The City in CinÂeÂma. FolÂlow him on TwitÂter at @colinmarshall or on FaceÂbook.



















