These “thinking” models get a rating from 1% to 1.3%. No ordinary model gained more than 1%. A prize of $ 700 thousand was announced for the creators of the first five models approaching the test.
The Arc Prize Foundation Foundation has released a new test to test AI’s capabilities compared to a person called ARC -i-2 and announced a prize of $ 700 thousand to the creators of the first five models that will receive 85% or higher ratings in this test. The information has been published on the fund website.
Unlike other trials to check AI for specialized knowledge or the ultra-flow of answers that are impossible for humans, arc -i-2 uses tasks that are simple for humans, but difficult for AI. In trials, each task was resolved by at least two people for two attempts (or fewer) in a controlled study involving four hundred people, so each AI model also received two attempts.
Fund researchers have found that artificial intelligence systems are difficult to interpret the meaning of the symbols, which exceeds their visual species. Systems have tried to evaluate symmetry, reflect glass, transform and even recognize connecting elements, but cannot determine the meaning of the symbols themselves. Also, II is difficult to accomplish tasks that require simultaneous application of policies or the application of multiple policies that interact with each other. If the policies of the tasks should be changed depending on the context, contemporary artificial intelligence will not be able to cope with the task.
It should be noted that despite the fact that the two selected people in the test groups always pass the test completely, the average result is one person – 60%.
We reminded, earlier reported that China announced the creation of the first autonomous agent of the world AI.
Cloudflare created AI to deceive bots collecting information for AI
News from CORRESPONDENT.NET On the telegram and whatsapp. Subscribe to our channels https://t.me/KorresPondentNet and WhatsApp
Source: korrespondent
I am Ben Stock, a passionate and experienced digital journalist working in the news industry. At the Buna Times, I write articles covering technology developments and related topics. I strive to provide reliable information that my readers can trust. My research skills are top-notch, as well as my ability to craft engaging stories on timely topics with clarity and accuracy.