• FavTutor
  • Posts
  • Testing AI Models with Minecraft Builds

Testing AI Models with Minecraft Builds

Including the latest AI news of the week

March 21, 2025

Hello, AI Enthusiasts!

Welcome to FavTutor’s AI Recap! We’ve gathered all the latest and important AI developments for the past 24 hours in one place, just for you.

In Today’s Newsletter: 😀

  • Testing AI Models with Minecraft Builds

  • 49% of Americans Believe AI Models Are Smarter Than Them

  • AI Scientist Writes Paper & Passes Peer Review Too

AI Research
🕹️ Testing AI Models with Minecraft Builds

There is a new popular website popped-up recently that lets you test and compare which AI models are good at Minecraft. There, users vote on the best Minecraft build made by AI Model’s prompts.

Insights for you:

  • MC-Bench or Minecraft Benchmarking compares two AI models on how well they can generate innovative Minecraft creations using the same prompt.

  • Here’s how it works: when you visit the website, you are shown two creations and you have to vote for which one looks better.

  • The website was created by a 12th-grader Aditya Singh with thousands of Minecraft builds ready to be evaluated right now.

AI Study
😢 49% of Americans Believe AI Models Are Smarter Than Them

Elon University’s Imagining the Digital Future Center recently did a survey revealing some interesting insights on how people are using and thinking about LLMs.

Insights for you:

  • The first finding was that 52% of American Adults are already using LLMs like ChatGPT, Gemini, or Copilot.

  • 49% of LLM users think that these models are smarter than them. 26% of the users even said they are a lot smarter.

  • ChatGPT is their favorite among the LLM users. 72% of them have used it.

  • Interestingly, most of the people are using it for their learning.

Sakana AI
🧪 AI Scientist Writes Paper & Passes Peer Review Too

AI Scientist is an innovative system designed to automate the entire scientific research process, made by Sakana AI. Recently, the company collaborated with a workshop at the ICLR (International Conference on Learning Representations) 2025.

Insights for you:

  • Sakana AI submitted 3 AI-generated papers to a workshop at ICLR 2025 to see if it can pass the peer review.

  • 1 out of the 3 papers surpassed the acceptance threshold. This means the AI-generated paper is equally good as human-authored research.

  • According to Sakana, this was the first time such a thing had happened: An AI-generated paper was reviewed and passed.