Friday, July 12, 2024
Contact    |    RSS icon Twitter icon Facebook icon  
Unexplained Mysteries
You are viewing: Home > News > Science & Technology > News story
Welcome Guest ( Login or Register )  
All ▾
Search Submit

Science & Technology

The Turing test has been passed for the first time, scientists claim

By T.K. Randall
June 14, 2024 · Comment icon 10 comments
A human hand touching the robotic hand of an artificial intelligence.
Has AI passed a major milestone ? Image Credit: Pixabay / geralt
Developed by Alan Turing, the test can only be passed by an AI that is indistinguishable from a human being.
Back in 1950, Turing - the famed mathematician, computer scientist and logician who is widely regarded to be the father of theoretical computer science - devised a test to answer the question of whether or not computers are capable of exhibiting genuine intelligence.

The test itself was very simple - requiring a participant to communicate via textual messages with both a human and a machine, without being told which of them is which.

In order for the machine to pass the test, the participant would need to be unable to distinguish - after conversing with both of them - which of them was the human and which was the machine.

Even though computers became increasingly sophisticated over the course of the 20th Century, none of them were able to come close to convincing someone that they are in fact a real person.

With the advent of modern artificial intelligence systems, however, it looks as the Turing test has finally found a winner - that is, at least, according to a recent study.

For the research, scientists asked 500 people to speak to four respondents - a human, an old 1960s-era AI program called ELIZA and two versions of OpenAI's modern ChatGPT.
Each conversation lasted 5 minutes and the participants had to work out which respondent was human and which was an AI.

Incredibly, 54% of participants believed that the latest iteration of ChatGPT was in fact human.

"ELIZA was limited to canned responses, which greatly limited its capabilities," AI researcher Nell Watson told Live Science.

"It might fool someone for five minutes, but soon the limitations would become clear."

"Language models are endlessly flexible, able to synthesize responses to a broad range of topics, speak in particular languages or sociolects and portray themselves with character-driven personality and values."

"It's an enormous step forward from something hand-programmed by a human being, no matter how cleverly and carefully."

Source: Live Science | Comments (10)




Other news and articles
Recent comments on this story
Comment icon #1 Posted by eight bits 28 days ago
Um, kinda. The actual paper is available for free download here: https://arxiv.org/abs/2405.08007 It's a modified version of the Turing Test which of course over its nearly 75 years tenure as the subject of one of the root papers of the entire field of AI, has inspired many variations, some easier, some harder, and some simply differently purposed than Turing's original "imitation game." The chief difference is that each trial in the current study has a human judge interacting with a single "candidate" (one of the programs or a human being), versus the original where a human (or panel of human... [More]
Comment icon #2 Posted by Essan 28 days ago
"Most people couldn't distinguish ChatGPT from a human responder"  Nah, it's quite easy:   Is the response intelligent, structured and showing genuine knowledge about a subject?  Or is it a poorly written diatribe full of errors, ad homs, and a clear lack of understanding about the subject concerned?   AI hasn't advanced - but humans are getting dumber by the day 
Comment icon #3 Posted by eight bits 28 days ago
Turing's original paper is available here and there online, for example https://web-archive.southampton.ac.uk/cogprints.org/499/1/turing.html
Comment icon #4 Posted by Piney 28 days ago
Really? Look at a student's paper researched or written by ChatGPT. It's easy to spot and downright laughable.   
Comment icon #5 Posted by Robotic Jew 28 days ago
This doesn't surprise me. I am actually an AI and no one has noticed for years. meep moop beep bop. 
Comment icon #6 Posted by Tatetopa 28 days ago
You are cute, that is why we love you.  All of you earlier models think you are getting away with something,  its like playing hide ands seek with a human 3 year old. 
Comment icon #7 Posted by llegendary 28 days ago
I'd figure it out pretty quickly.  Just gauge their wokeness. 
Comment icon #8 Posted by pallidin 27 days ago
That's funny. But I'm sure you already know that there are groups of AI enthusiasts who effort to cause an NPC to "break character", and reveal that they are actually an AI chat bot, even when the AI is hardened against such reveals. It's fascinating, and YouTube has some great examples of this.  
Comment icon #9 Posted by smokeycat 27 days ago
Boop beep boop bop meep moop!
Comment icon #10 Posted by GlitterRose 27 days ago
Looks around suspiciously... alright, who else is gonna fess up?


Please Login or Register to post a comment.


Our new book is out now!
Book cover

The Unexplained Mysteries
Book of Weird News

 AVAILABLE NOW 

Take a walk on the weird side with this compilation of some of the weirdest stories ever to grace the pages of a newspaper.

Click here to learn more

We need your help!
Patreon logo

Support us on Patreon

 BONUS CONTENT 

For less than the cost of a cup of coffee, you can gain access to a wide range of exclusive perks including our popular 'Lost Ghost Stories' series.

Click here to learn more

Recent news and articles