(description::A draft of the blog post has to be published on Bearblog. )
Currently, both the human brain and AI systems operate largely as “black boxes.” 🧠 Notably, AI training architectures, especially neural networks, were significantly influenced by our understanding of brain function. Modern AI/LLMs are comparable to a child: they learn by replicating speech and text, not through genuine comprehension. This indicates AI is in a nascent state, projected to mature considerably within 20-30 years.
Intriguingly, insights gained from AI training could potentially unlock the mysteries of our own brain’s functioning. This enhanced understanding of the brain might, in turn, catalyze further breakthroughs in AI technology itself.