32 Times AI Didn't Understand The Assignment And Created Hilariously Weird Images

We got in touch with the creator of 'Weird AI Generations' and they are very optimistic about the technology's prospects in the coming years.

"I expect AI images to produce even better quality in the future," they told Bored Panda. "In my opinion, they are going to be much more in-depth and more accurate to the request, too."

Some challenges, however, might be ingrained in the technology itself. Ahmed Elgammal, a professor of computer science and director of the Art & AI Lab at Rutgers University, agrees that newer versions of generative AI will give users more control over outputs, along with higher resolutions and better image quality.

However, Elgammal said that "the main limitation, as far as art is concerned, is foundational: it's the process of using language as the main driver in generating the image."

"Visual artists, by definition, are visual thinkers," he explained. "When they imagine their work, they usually draw from visual references, not words – a memory, a collection of photographs, or other art they've [already] encountered."

When language is in the driver’s seat of image generation, there might be an extra barrier between the artist and the digital canvas. If pixels are rendered only through the lens of language, artists might have to revamp their creative process, as they will become constrained within the boundaries of semantics.

Half joking and half serious, the creator of the Facebook page also pointed to what seems to be one of generative AI's biggest current problems, stating, "For now, I believe the images would appear much more realistic if they accurately depicted fingers."

Seyedali Mirjalili, director of the Centre for Artificial Intelligence Research and Optimisation at Torrens University, explained that this is due to the training images, in which hands are often small, holding objects, or partially obscured by other elements.

"It becomes challenging for AI to associate the term 'hand' with the exact representation of a human hand with five fingers," Mirjalili added.

But the researcher thinks it's only a matter of time before the technology will move past this.

"It's important to remember text-to-image and text-to-video conversion is a relatively new concept in AI," he said. "Current generative platforms are 'low-resolution' versions of what we can expect in the future."

"With advancements being made in training processes and AI technology, future AI image generators will likely be much more capable of producing accurate visualizations," he noted.