minus-squareJanx@piefed.socialtoTechnology@lemmy.world•Announcing ARC-AGI-3 - A benchmark that tests if AI can explore, learn, and adapt in unfamiliar situations. Humans score 100%. Frontier AI scores 0.26%.linkfedilinkEnglisharrow-up3·3 days agoGrok isn’t designed to solve problems. It’s designed to create sexually explicit images of children for Republicans… linkfedilink
Grok isn’t designed to solve problems. It’s designed to create sexually explicit images of children for Republicans…