Hacker News Reader

I totally agree with this. In spite of a lot of the recent hype, we still have very limited understanding of intelligence as a science. I agree "never" seems like a strong claim given that we're walking around with 7B existence proofs that it's possible, but for most of us "never" and 300 years are functionally equivalent.

View on HN

What makes you think that it would take 300 years, given that humans are being beaten at extremely creative computational tasks like playing Go, already? Machine learning using similar architecture (neural nets) seems to be making astounding progress every day. Why would it take 300 years to make a major breakthrough?

We've made big advances in reinforcement learning and certain types of supervised learning. But most learning is unsupervised. Somehow a human baby bootstraps itself from being a tangle of neurons to sensing, walking, and talking with very little instruction on how to do so.

But monkeys don't start talking (even when surrounded by humans), therefore would you agree that somewhere in our DNA must be the description for how our neurons need to do so? Therefore, that DNA must contain this information? (source code), at some level of abstraction or compressed in some way?

Reply on news.ycombinator.com