arxiv.org/abs/1911.01547 “On the measure of intelligence” where he proposes a new benchmark for “intelligence” called the “Abstraction and Reasoning corpus”.
Highlights below ->
![](https://pbs.twimg.com/media/EKaz_67UwAAqCdR.png)
![](https://pbs.twimg.com/media/EKa0i5hU4AAjRtz.jpg)
![](https://pbs.twimg.com/media/EKa0u4oUwAAXpTw.png)
![](https://pbs.twimg.com/media/EKa1twkU8AAKLkS.png)
![](https://pbs.twimg.com/media/EKa2HbLUcAExNFA.png)
![](https://pbs.twimg.com/media/EKa3m1-UUAAKkff.png)
![](https://pbs.twimg.com/media/EKa5SRUU8AA1i_4.png)
![](https://pbs.twimg.com/media/EKa5ao6UYAAEz9S.png)
![](https://pbs.twimg.com/media/EKa5mgFVAAA3G8X.png)
I find it incredibly easy for humans but i have no idea how to program something that would solve it without being super-specialized...
![](https://pbs.twimg.com/media/EKa6eZuU8AIG8tW.png)
- They are relatively easy for humans
- The performance of any known AI system on them is close to zero
Which means that there is only one way to go for AI: UP!
🥳
![](https://pbs.twimg.com/media/EKa7sPaU0AEpuUn.png)
github.com/fchollet/ARC
Git clone the repo and open the HTML interface in a browser. It’s a lot of fun to play with! 😃