AI Ulysses – Visually

A sister program from OpenAI to ChatGPT which I’ve experimented with is Dall-E2. This does for images what the other program does for text. In response to prompts, it creates realistic images and art. Just as knowledge workers should find out about ChatGPT before it finds them, so graphic designers and illustrators should check out Dall-E2.

I experimented with Ulysses, supplying snippets of the poetic text that stood out for me when I read the novel. Here’s some of the results.

**UPDATE: I ran the same prompts through Midjourney and added the results below. Some are the original 4x results, others were upscaled.

Dalle2: “Old wall where sudden lizards flash.” (Scylla and Charybdis, p. 194)
Midjourney: “Old wall where sudden lizards flash.”
Dalle2: “Where fallen archangels flung the stars of their brows.” (Wandering Rocks, p. 232)
Midjourney (4x): “Where fallen archangels flung the stars of their brows.”
Dalle2: “…stars hung with humid nightblue fruit.” (Ithaca, p. 651)
Midjourney (4x): “…stars hung with humid nightblue fruit.”
Dalle2: “…the extreme boundary of space…” (Ithaca, p. 680)
Midjourney (4x): “…the extreme boundary of space…”
Dalle2: “…past beds of reeds, over slime, mudchoked bottles, carrion dogs.” (Hades, p. 95)
Midjourney: “…past beds of reeds, over slime, mudchoked bottles, carrion dogs.”

1 Comment so far
Leave a comment

Weekend Financial Times arts columnist Jan Dalley comments that this software (with the similar name “all that is missing is a ‘Y'”)

…has sent ripples, if not shockwaves, through a community of creative people: does it challenge their reason for being and our notions of what art should be?
More seriously, what does this mean for old-school notions of copyright and intellectual property? Does Hockney hold any rights over turquoise pools with a stripy undertow? Does Hirst own spot patterns? Obviously not — but still we know a rip-off when we see it. And there are aggrieved yelps across the digital world from creators who know that massively profitable sites are built by scraping the internet for their original work.

These issues have been gnawed over ever since this kind of software first emerged. With Dall-E-2, the law is already clear: “in the eyes of the Copyright Office, the public is free to reproduce, publish, or sell” any old piece of rubbish you create — whoops, sorry, “your Dall-E 2-generated masterpiece” — with no strings attached. Its creators freely admit that the GAI has been “trained”, as I’ve learnt to call it, with copyright images and original work that might come under all sorts of restrictions, but there’s no actual crime here, it seems.

Possibly not. The debates continue. What surprises me most about the back-and-forth around the subject is a pervasive view that GAIs are capable of creating beautiful and skilful images that might actually fool someone, or replace an artist’s real work. If that’s true, I must have been looking in the wrong place.

Leave a comment
Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>



four × 4 =