Posted on 07/06/2023 5:25:02 AM PDT by FarCenter
New research from the University of Montana and its partners suggests artificial intelligence can match the top 1% of human thinkers on a standard test for creativity.
The study was directed by Dr. Erik Guzik, an assistant clinical professor in UM's College of Business. He and his partners used the Torrance Tests of Creative Thinking, a well-known tool used for decades to assess human creativity.
The researchers submitted eight responses generated by ChatGPT, the application powered by the GPT-4 artificial intelligence engine. They also submitted answers from a control group of 24 UM students taking Guzik's entrepreneurship and personal finance classes. These scores were compared with 2,700 college students nationally who took the TTCT in 2016. All submissions were scored by Scholastic Testing Service, which didn't know AI was involved.
The results placed ChatGPT in elite company for creativity. The AI application was in the top percentile for fluency -- the ability to generate a large volume of ideas -- and for originality -- the ability to come up with new ideas. The AI slipped a bit -- to the 97th percentile -- for flexibility, the ability to generate different types and categories of ideas.
"For ChatGPT and GPT-4, we showed for the first time that it performs in the top 1% for originality," Guzik said. "That was new."
He was gratified to note that some of his UM students also performed in the top 1%. However, ChatGTP outperformed the vast majority of college students nationally.
(Excerpt) Read more at sciencedaily.com ...
Top 1%?
That means one of them cheated!....................
Ask ChatGPT if the 2020 Presidential election was fraudulent.
There’s nothing original about AI’s responses, when everything ‘it’ knows has been programmed into it by humans.
All of a sudden, poorly-proctored tests for creativity result in 100% of test-takers falling within the top 1%.
AI is programmed but also trained.
Once programmed, an AI is fed a large mass of data from which the program adjusts a vast number of internal variables. The training set for general purpose AI is usually obtained by reading the internet. One source may be the 6.5 million articles in Wikipedia.
As a result, the AI can perform tasks that were not envisioned by the AI’s programmers.
“AI” is not intelligent. “AI” is not an oracle. “AI” is a record storage facility for patterns.
Was this research conducted by AI?
Ask a generative “AI” to tell a reative story with the freedom cranked up and you typically get a chaotic mis-mash of tropes twisted in some clueless way that makes little overall sense. It can be entertaining for a while to giggle at the silliness of it.
No so.
It learns. And it can actually write programs and then insert the programs it has written into its own code.
You don't actually think scientists would actually be afraid of AI if it was limited to what was programmed into it?
I write routines to analyze stocks and use ChatGPT (the most basic model) to write and debug code in VBA for Excel. It significantly improves the speed and accuracy of my work. When I ask it a question for a unique routine, no programmer working on ChatGPT has anticipated my question and previously input this program into it.
This simple AI system reads my question, learns what I am seeking, and responds within a matter of seconds with some pretty complicated programs. It learns.
Does spewing out off-the-wall nonsense count as creativity?
Pattern matching is one of the hallmarks of a high IQ.
Who decides what qualifies as “original creative thinking”? The leftists who programmed the AI?
To what use?
“As a result, the AI can perform tasks that were not envisioned by the AI’s programmers.”
You need to explain. The tasks that the computer performs are, by definition, programs that are done by humans.
Yes, a program could generate other programs but the computer-generated programs are the output of human-created programs.
The ultimate, simplest AI thing is a random number generator.
Keep that hype bubble going, gotta get those investors dollars in our back pocket before the fall...
Investors never learn.
AI isn’t AI...
Probability algorithms are not “thought”.
They have their uses, but the hype around this stuff is just laughable right now... basically its is the current “self driving car” hype.
It is interesting, it will change some things, like all tech does, but its not remotely going to be what they are hyping it out to be at the moment.
One has to consider that the control group was University of Montana students ;-)
> Probability algorithms are not “thought”.
But they’ll beat the judgement of most people.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.