AN ARTIFICIAL intelligence program has developed its own language and no one can understand it.<\/p>\n
OpenAI is an artificial intelligence systems developer - their programs are fantastic examples of super-computing but there are quirks.<\/p>\n
3<\/p>\n
3<\/p>\n
3<\/p>\n
DALLE-E2 is OpenAI's latest AI system - it can generate realistic or artistic images from user-entered text descriptions.<\/p>\n
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.\"<\/p>\n
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.<\/p>\n
But the system has one strange behavior - it's writing its own language of random arrangements of letters, and researchers don't know why.<\/p>\n
Giannis Daras, a computer science Ph.D. student at the University of Texas, published a Twitter thread detailing DALLE-E2's unexplained new language. <\/p>\n
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.<\/p>\n
Daras fed \"vicootes\" back into the DALLE-E2 system and got back pictures of vegetables.<\/p>\n
\"We then feed the words: 'Apoploe vesrreaitars' and we get birds.\" Daras wrote on Twitter. <\/p>\n
\"It seems that the farmers are talking about birds, messing with their vegetables!\"<\/p>\n
Daras and a co-author have written a paper on DALLE-E2's \"hidden vocabulary\". <\/p>\n
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\".<\/p>\n
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.<\/p>\n
Some AI researchers argued that DALLE-E2's gibberish text is \"random noise\".<\/p>\n
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.<\/p>\n
We pay for your stories!<\/p>\n
Do you have a story for The US Sun team?<\/p>\n
<\/p>\n
View original post here: <\/p>\n
Artificial intelligence spotted inventing its own creepy language and its baffling researchers... - The US Sun<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" AN ARTIFICIAL intelligence program has developed its own language and no one can understand it. Continue reading