AN ARTIFICIAL intelligence program has developed its own language and no one can understand it.

OpenAI is an artificial intelligence systems developer – their programs are fantastic examples of super-computing but there are quirks.

An AI system has assigned these gibberish names to objects and researchers don't know why

1

An AI system has assigned these gibberish names to objects and researchers don’t know why

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.

AI 'face search engine' sparks fears as its finds pictures of you on the web
Could artificial intelligence REALLY wipe out humanity?

Giannis Daras, a computer science PhD 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.

Daras fed “vicootes” back into the DALLE-E2 system and got back pictures of vegetables.

“We then feed the words: ‘Apoploe vesrreaitars’ and we get birds.” Daras wrote on Twitter. “It seems that the farmers are talking about birds, messing with their vegetables!”

Most read in Tech

Daras and a co-author have written a paper on DALLE-E2’s “hidden vocabulary”.

They acknowledge that telling DALLE-E2 to generate images of words – the command “an image of the word airplane” is Daras’ example – normally results in DALLE-E2 spitting out “gibberish text”.

When plugged back into DALLE-E2, that gibberish text will result in images of airplanes – which says something about the way DALLE-E2 talks to and thinks of itself.

'Mass shooting' at funeral leaves 'at least five hurt' as cops ask to avoid area
Mama June's daughter Pumpkin, 22, shows off bump in photoshoot with sisters

Some AI researchers argued that DALLE-E2’s gibberish text is “random noise“.

Hopefully, we don’t come to find the DALLE-E2’s second language was a security flaw that needed patching after it’s too late.

We pay for your stories!

Do you have a story for The US Sun team?

This post first appeared on Thesun.co.uk

Leave a Reply

Your email address will not be published. Required fields are marked *

You May Also Like

Roman soldier’s payslip from 1,900 years ago reveals he was left BROKE after military deductions

A payslip made from a sheet of papyrus shows a Roman soldier…

Will Microsoft’s acquisition of Activision Blizzard finally bring scrutiny on the video game industry? | Akin Olla

The video game industry is bigger than the film and music industries…

It’s Time to Focus on Reproductive Longevity Research

Women’s ovaries age prematurely, at more than twice the rate of other…

Google Bard and SGE AI accused of ‘promoting benefits of fascism, and genocide’ and said ‘slavery was good for economy’

CONCERNED tech experts have claimed they experienced interactions with Google’s AI bots…