Cory Dransfeldt has great posts about AI, and this one is no exception. Of course, he’s right in his stance on AI-generated images, but as with everything, I don’t find the conclusion so simple.

The images it generates are, at best, a polished regression to the mean. If you want custom art, pay an artist.

As I have covered before, my thoughts on generative AI are mixed. While I understand many of the issues people have with it, I can’t find the same motivation. I use it many times a day to help with tasks that would have taken me much longer; it helps me out with code issues I have, and I also use it to generate images for my blog posts. While I don’t do this a lot, these generated images are a replacement for using stock images. Photos from places like Unsplash are not uncommon to see on the web, pushed by platforms like Ghost and WordPress, and these do not generate any income for artists either.

I would love to have the money to pay an artist to produce images for my blog, but this is not viable, which I am sure is the case for many people. Generative AI just helps me out now and again to make my posts a little more appealing. That is not to say I don’t take onboard the ethical issues; my use of these images has plummeted in the last few months due to this, and in many instances where I would have used an image in the past, I just publish the post with text.

There should be a safe middle ground to cater to these ethical conundrums, as there should be for stopping AI from hoovering up your data - but I just don’t see one happening.