A glimpse of an awesome new kind of intelligence


One Giant Step for a Chess-Playing Machine

In this quote, Stockfish is a devastatingly effective, but traditional, chess-playing program that easily beats humans:

Most unnerving was that AlphaZero seemed to express insight. It played like no computer ever has, intuitively and beautifully, with a romantic, attacking style. It played gambits and took risks. In some games it paralyzed Stockfish and toyed with it. ... Grandmasters had never seen anything like it. AlphaZero had the finesse of a virtuoso and the power of a machine. It was humankind’s first glimpse of an awesome new kind of intelligence.

Coincidentally, this article also dovetails with Homo Deus by Yuval Harari, which I mentioned in my last post. We are used to thinking of consciousness and intelligence as inseparable, because in humans, they do go together. Though there's now lots of evidence that machines can be intelligent, there is no evidence as yet that they are conscious. Lacking that, it's meaningless to talk about what machines want. They do whatever we program, or in this case train, them to do. And we must be cautious about anthropomorphizing. When the article's author writes that AlphaZero "toyed with" its opponent, that makes us think of an interaction with emotions attached, like "playfulness" or "glee". Banish that thought: these machines aren't built to have emotions.

Some good stuff in that book (I guess this is turning into a strong endorsement) about where AI might be taking us. One distinct possibility is that we start turning over more of our decision-making to the machines, because they are simply better at it than we are capable. You may have heard that algorithms analyzing Facebook "likes" are better, after gathering 300 data points or so, at knowing us than we know ourselves. So yes, we might end up having blindly to trust the "AlphaInfinity" this article posits because it's always right, even when we're unable to understand why.

Not discussed in this article is another distinct possibility: that we (or some of us) begin upgrading our own mental hardware so we can keep up with the AIs. That might save us from living in an incomprehensible world. Then again, it might initiate a new sort of arms race, an intelligence race, since significantly smarter people would hold a big advantage over everyone else.

One other small point: the article's author starts out talking about "AlphaZero" and ends with speculating about "AlphaInfinity". But to me, "AlphaZero" sounds suspiciously close to "Aleph Zero"... which is another way of saying "infinity". Maybe Google/Alphabet Inc. was hinting at the infinite possibilities of their system, already.

Original Facebook post

Comments

Post a Comment

Popular posts from this blog

18 Lessons Learned After 18 Years At Amazon

Image
I’ve been at Amazon as a software developer continuously for over 18 years now. That’s longer than most. Lately I’ve been reflecting on my time there. Here are 18 lessons I've learned -- most of which are applicable to any collaborative endeavor. Trust is your most important asset. It’s slow to build and easy to lose. This is true inside the company (with teammates, reports, and managers) and outside (with customers, partners, and investors). Error correction is at the heart of development. Humans make errors, and a lot of humans make a lot of errors. This is unavoidable, and one should never blame the person – always the process. Swift and honest error correction builds trust. This, too, is true inside the company (with teammates, reports, and managers) and outside (with customers, partners, and investors). Transparency is the best policy. If you don’t have enough informa...
Read more

Being Intelligent, Whatever That Means

Image
This post was inspired by a talk called Being Human , given by Jason Wodicka, someone I worked with at Amazon. This is a great talk, and you should watch it. Jason is a smart guy, and while he and I may not have arrived at exactly the same conclusions (or perhaps we "violently agree" and are just looking at similar ideas from different directions), I absolutely appreciate where he is coming from. What follows are my personal reactions and responses to the talk. Also I should note that this is hardly my first blog post on AI; if you'd like to read the others, see Never Is A Long Time and Buckle Up and... well, any of these . Yes, Jason: You're quite right about the boom-and-bust, summer-and-winter cycle of AI. It's looked promising before, and we've been disappointed before. I, too, have been around for a while, and I too used to read Wired magazine in print. 😄 We must approach reputed AI developments with a big ol' grain of salt. Turing's "imi...
Read more

Here, There Be Dragon... Rubies...

Image
Big software projects, at big software companies, are deeply collaborative affairs. The majority of the work isn't writing code; it's getting a shared vision into everybody's heads. That's not what this post is about, though. Since I decided to take a little break from my day job and decompress, I found an activity that relaxes me, and reminds me of my earliest days writing software. First I'd like to describe what my earliest paid software gig was like. Then I'll talk about a personal project I worked on back in 2012, and then another that I'm working on now. Memory Lane: QANIMATE Back in my High School days, I worked on a project called CUPS, "Consortium of Upper-level Physics Software". You can still find traces of it online . The piece I wrote is QANIMATE , although they did misspell my last name in the author attribution! I worked with Ron Stoner at BGSU  over the course of a summer. Essentially it was an internship, an educational opportunity...
Read more