Sumbetter is a platform where human and AI language creativity can be compared.
There is a rewording competition on the platform. We list 8 interesting quotes that can be paraphrased by anyone who has registered with a pseudonym. We post these paraphrased sentences on the platform. In the meantime, we also paraphrase the quotes with LLMs and post them under invented pseudonyms. But whoever reads the paraphrased texts does not know whether it was done by a model or a human.
Before posting a text, registered users should rate 5 texts from 1 to 5 stars. (There would be a button to report foul language.)
The competition would take place in several languages, so the nations would compete with each other.
We would announce results after 3 months. We would see if AI or humans are more creative in wording.
My hypothesis is that human creativity surpasses that of LLMs. And my question is, is there any difference between languages in how many ways they can express the same idea. The result is to confirm or reject the hypotheses.
10 000 USD for building the platform.
10 000 USD for online marketing activities.
9 000 USD for the work I will do for 6 months. (This is a full-time job.)
I am a sole founder.
A few years ago, I planned and managed an inter-company competition for the Count Széchenyi Family Foundation, where we measured and awarded companies' innovation ability, productivity and CSR activities.
It fails, if people do not want to prove that their language is wonderful. It fails if we can't find those people who think that they have got talent to languages.
If it fails, we lose much money and an illusion about humanity.
0.
App status across various funders