Naturally the first comments here are criticisms that the characters change from frame to frame, but I think that's worth putting aside, since any rational person will understand that systems like DALL-E will have the ability to maintain a level of continuity between images in the near future (if it's not already there and just not exposed).

Besides that, this is pretty good - certainly a few things that don't seem quite ideal (the fifth frame doesn't seem to capture the meaning of the text especially well), but enough to make you think that in not too long, it is fully plausible that DALL-E will be fully able to act as an illustrator for this use case. On the one hand, pretty exciting. On the other, certainly harrowing for children's book illustrators.

I wonder how long it will be before I can use this for my marketing emails. I sell dog treats, and I have a fairly simple template with an image at the top. That's almost always a photo of my products and/or my dogs. How long before I can just ask for an image from DALL-E for something like "Dog sitting next to grill in back yard with American flags and other patriotic decorations" for the top of my Fourth of July email? How long before I can feed it a picture of my products and have it generate photorealistic images of dogs eating them, thus replacing the photographer I use for product shoots? It's an exciting prospect for me as a small business owner - lots of time and money saved in an area where I'm not an expert - but definitely pretty scary for people who create visual imagery of any kind.