Applications using artificial intelligence to create content seem to be all the rage today.
One could be forgiven for brushing off such automation as a fad, but sometimes these technologies deliver some very interesting results.
One example is the mobile phone application WOMBO Dream, which takes images from the Internet to create composites. All it requires is typing in a text prompt.
Feeling curious about what the results would be if we tried out this technology with keywords related to Yucatán, we promptly got to work. Here is what we got.
As you can see, the results of this “AI art” are extremely figurative but have an appealing allure.
Whether or not these pieces should really be referred to as art is ultimately a philosophical question, but they are certainly interesting to look at.
The AI images evoke their alleged subject matter in a way that is simultaneously quite literal and bizarre.
As the exact phrasing of the text inputs has big effects on the resulting image, we experimented with several spellings in both English and Spanish, as well as Mayan. Even case sensitivity had an effect on the results.
Though during our tests inputting the same text twice produced the same image, this may not always be the case as the graphics used by the system are likely to update from time to time.
None of the images have been manipulated or changed in any way, other than some cropping for size consistency.
Almost all of the images produced by the AI came out being quite colorful, which may be a feature of the system or simply the result of the prompts it was given.
One of the few darker-looking images. We assume the figure at the top center of the image is supposed to be the Monument to the Montejo.
We also experimented with some more generic Mexican terms.
Source: WOMBO Dream