Sunday, March 27, 2016

Issues of AlphaGo's victory

In my last post, "Victory for AI", I wrote about the computer program AlphaGo and its victory in a game of Go with Lee Se-dol of South Korea, the world champion of the game. So many news sites, blogs and websites have written about this news, so there's plenty for ESP teachers to choose from to use for discussion, technical description, or focus on artificial intelligence.

One of the main areas focused on is the question of the impact this development could have - will AI eventually become smarter than humans and take over the world. Many articles see this as a possibility, while others feel this is an unlikely scenario.

An article in the Sunday Times (Great Britain) presents some of the issues, and ends with the writer's opinion. The article is, "I for one welcome the rise of the robots. They can do the work while I play."

Link: http://www.thesundaytimes.co.uk/sto/comment/columns/dominiclawson/article1679638.ece

Some of the points mentioned:
  • "why Go has been seen as such a good testing ground for the holy grail of modern computing, artificial intelligence"
  • "Hassabis, however, seems most gripped by what it could do to solve humanity's most pressing anxieties."
  • "Will these programs develop to the stage when most, if not all, of our professions are made redundant?"
  • "Others worry most about the military applications of artificial intelligence."
  • "And what if such technology got into the wrong hands?"
  • The author quotes Stephen Hawking: "The primitive forms of artificial intelligence we already have have proved very useful. But I think the development of full artificial intelligence could spell the end of the human race."
  • And the author's opinion: "And would it matter if the existing skills of the rest of us were largely rendered redundant by his work? If they were, it would mean that AI had given the world a superabundance of what it needs to live comfortably, at a very low cost."


For a focus on some of the ethical issues involved, I checked one of my favorite blogs, Engineering Ethics. I wrote about the blog in my post of January 14, 2015 ("Relevant links"), and there is a link to it under Relevant Links on this blog. The post is titled, "AlphaGo Deeats Human Go Champion: Go Figure."

Here is the link to this specific post:  http://engineeringethicsblog.blogspot.co.at/2016/03/alphago-defeats-human-go-champion-go.html

The blogger refers to the article in Wired magazine that I mentioned in my last post. He is also of the opinion that intelligent machines are unlikely to become more powerful than humans. At the end of his post he comments:

"And as long as enough people remember than humans are not machines, and machines are made by, and should be controlled by, humans, I think we don't have to lose a lot of sleep about machines taking over the world. What we should watch are the humans running the machines." (emphasis is mine)

I think this point is an important one, and it has come up often in discussions about issues in engineering ethics that I've had with my students.

I also looked at the blog Robotics and AI - also on my Relevant Links - since this is an area the blog particularly focuses on. Their article is "Machine vs. Man. What happened when AI played against a World champion."

Link to post:  http://www.roboticsandai.com/2016/03/machine-vs-man-what-happened-when-ai-played-against-a-world-champion/

This post gives a general overview of the news and, not surprisingly, concludes with:

"one thing is clear ... machines with intuitive capabilities are a not too distant reality."

Whether this will be a good thing or not -- well, that's an interesting topic to discuss with our students.

No comments:

Post a Comment