The world of technology after an artificial intelligence programmed reached the pinnacle of independence by writing its own language. Nobody else is capable of fully understanding the language coined by Open AI's "DALLE-E2" artificial intelligence system.

Figure 1: Open AI's "DALLE-E2" artificial intelligence system.

Figure 1 shows that Its job is to generate realistic and/or artistic images based on text descriptions entered by users. Open AI claims that DALLE-E2 is groundbreaking, for it effectively "learned the relationship between images and the text used to describe them." [1]

DALLE-E2 is OpenAI‘s latest AI system – it can generate realistic or artistic images from user-entered text descriptions.

DALLE-E2 represents a milestone in machine learning – OpenAI’s site says the program “learned the relationship between images and the text used to describe them.”

A DALLE-E2 demonstration includes interactive keywords for visiting users to play with and generate images – toggling different keywords will result in different images, styles, and subjects.

But the system has one strange behavior – it’s writing its own language of random arrangements of letters, and researchers don’t know why.

Giannis Daras, a computer science Ph.D. student at the University of Texas, published a Twitter thread detailing DALLE-E2’s unexplained new language.

Daras told DALLE-E2 to create an image of “farmers talking about vegetables” and the program did so, but the farmers’ speech read “vicootes” – some unknown AI word. [2]

Daras and a co-author wrote an article about the “hidden vocabulary” of DALLE-E2. They recognize that telling DALLE-E2 to generate images of words – the command “an image of the word airplane” is Daras’ example – typically results in DALLE-E2 spitting out “nonsensical text”.

When plugged back into the DALLE-E2, this nonsense text will result in images of planes – which says something about the way the DALLE-E2 talks and thinks about itself. Some AI researchers have argued that the DALLE-E2’s nonsense text is “random noise”.

Fortunately, we didn’t discover that the second language of DALLE-E2 was a security flaw that needed to be fixed after it was too late. [3]

References:

1. https://www.indiatimes.com/technology/news/artificial-intelligence-caught-writing-its-own-creepy-language-by-researchers-571452.html
2. https://nypost.com/2022/06/03/artificial-intelligence-spotted-inventing-its-own-creepy-language/
3. https://zhivoemilo.com/artificial-intelligence-caught-inventing-its-own-scary-language/

Cite this article:

Sri Vasagi K (2022), AI Developed Its Own Language, AnaTechMaz, pp.115