By now, the whole world has heard about Willy’s Chocolate Experience in Glasgow (archived), or as the page title has it, “Willy Choclate Experience”, which sounds like something quite different and not at all for the weans. That mangled English is in keeping with the AI-generated graphics promising “a pasadise of sweet teats”, “enigemic sounds” and “ukxepcted twits”, which is the usual xepctation in the uk these days.
Berthe Morisot comes into her own (via Mefi). I’ve seen a few of Morisot’s paintings in galleries, and they were always as good as the other Impressionist canvases around them; I guess I assumed that she hadn’t painted much, and that that was why there weren’t more on show. Learning that “an astounding proportion of [her] most important work” is still in private hands explains a lot, and sexism would explain the rest—what’s the bet that even some of her paintings in public collections are sitting in storage rather than being on display. It’s good to learn that her peers were so supportive of her work, and celebrated it after her death—the fault lies with posterity, but fortunately that can change.
A British Museum video about the artist Yamamoto Akane has to be one of my favourite links posted to Metafilter in months. What a stunning amount of work is conveyed in those few minutes; given the failure rate, the single object shown there must represent months of careful effort. How many of us could bear to start from scratch again and again when firings go awry?
So many of the artworks in the great museums and galleries of the world represent a comparable amount of painstaking and patient work, and are the result of years of learning, developing or—as here—inventing one’s craft. And most of us take them in during a visit in an instant and move on to the next.
Getting ChatGPT to write for you is all the rage nowadays, so I asked it to “write a blog post in a conversational style about starting the new year after returning to Edinburgh from Australia, and referring to a photo you took from a window seat on a flight over northeast USA at night”. Here’s what it burped out…
I’ve been continuing my earlier experiments with DALL-E 2 and Stable Diffusion—now with added Midjourney. Click through for a gallery.
Since my initial burst of activity with DALL-E 2 I’ve tried a few more queries on it, most of which didn’t produce much of interest, but there were a few exceptions. Things got more interesting when I was able to compare it with Stable Diffusion, which has been opening up to beta testers and let me (and thousands of others) on a few days ago. Unlike DALL-E 2, Stable Diffusion is aiming to be completely open, with no holds barred, which means it’s inevitably going to be used to churn out AI-generated porn and to rip off living artists. Not the greatest advertisement for the power and promise of artificial intelligence. But judging from some of its results, I wouldn’t panic just yet (click through for a gallery)…
It seems a lifetime ago now, but as an undergraduate I took courses on Computer Graphics, Digital Image Processing and Artificial Intelligence, so it’s no surprise that I’ve been fascinated by DALL-E 2 and the results that people have been getting from it. I signed up for the wait-list two months ago, and yesterday received an invite. I’ve already burned through half of my fifty credits for the month, so had better start pacing myself… but it’s so much fun. Here’s a gallery of my best results so far.
