{"id":24296,"date":"2015-07-14T21:01:11","date_gmt":"2015-07-14T13:01:11","guid":{"rendered":"https:\/\/www.hongkiat.com\/blog\/?p=24296"},"modified":"2023-04-17T21:42:19","modified_gmt":"2023-04-17T13:42:19","slug":"understanding-artificial-intelligence","status":"publish","type":"post","link":"https:\/\/www.hongkiat.com\/blog\/understanding-artificial-intelligence\/","title":{"rendered":"Understanding Artificial Intelligence – and Why We Fear It"},"content":{"rendered":"

Whether we admit it or not, we have a serious fixation with trying to create machines that can think for themselves<\/strong> \u2013 machines imbued with artificial intelligence, or AI, for short. In fact it is often the source of inspiration for the movie and TV shows that we watch: 2001: The Space Odyssey,<\/em> Spielberg\u2019s A.I, <\/em>the Terminator <\/em>franchise, Her, <\/em>Ex Machina, I, Robot, The Big Bang Theory etc<\/em><\/p>\n

Lately AI has been making waves again albeit it is of the ominous kind. Researchers are making tons of breakthroughs. The world\u2019s most powerful governments and military forces are investing heavily in it, and it has become something so tangible that the world\u2019s brightest minds have found themselves divided about it.<\/p>\n

While the likes of Bill Gates, Stephen Hawking, and Elon Musk are wary about AI, Yahoo! CEO Marissa Meyer, researchers at Google, and the public at large are absolutely thrilled by the idea. We take a look at what the hullabaloo is all about.<\/p>\n

Smart Devices vs AI<\/h2>\n

Any smart device can become \u201csmarter\u201d with the right app (like what you see here<\/a>). For example, you can turn your smartphone<\/a> into a key for your car, home or even a sensor with the right app and and\/or an accompanying gadget. However, these do not qualify as \u201cintelligence\u201d at all \u2013they are merely a set of instructions that rely on human input (instructions) in order to come up with an output.<\/p>\n

\u201cArtificial intelligence\u201d was coined by Stanford\u2019s John McCarthy at a conference in 1956. Artificial intelligence involves developing computers that can \u201cthink\u201d independently<\/em> \u2013 computers that act intelligently <\/strong>and perform functions<\/strong> normally done by man. But more importantly AI is a technology that learns by itself.<\/p>\n

In August 2014, news report about an AI that learns like a toddler<\/a> made rounds in the tech news world. It can learn in real-time and can mimic people\u2019s facial expressions. It also \u201clooks\u201d like a toddler \u2013 and I\u2019m not sure if that makes it less or more creepy.<\/p>\n

<\/figure>\n
\n
<\/i><\/div>\n
IMAGE: Auckland Bioengineering Institute Laboratory for Animate Technologies<\/div>\n
<\/div>\n<\/div>\n

AI Developments in the Field<\/h2>\n

There is a wide assortment of apps out in the market today that make the most out of what we\u2019ve been able to accomplish with AI so far \u2013 from pathfinding apps to personal assistants<\/p>\n

Cortana<\/a>, Microsoft\u2019s answer to Siri<\/a>, made use of its research on speech and language, allowing the app to participate in back-and-forth exchanges with users<\/strong> on their mobile phones. IBM has Watson<\/a>, a powerful intelligent system which became popular for beating a Jeopardy! champion in 2011.<\/p>\n

Google is also working hard on \u201cdeveloping algorithms<\/a> with the capacity for logic, natural conversation and even flirtation.\u201d The algorithms, called \u201cthought vectors,\u201d are supposedly being geared towards achieving a level of reasoning and logic comparable to that of human beings<\/strong> \u2013 AI with common sense, if you will.<\/p>\n

Then, there is Emily Howell. \u201cShe\u201d is an AI that has a knack for composing music<\/a> and generating controversy. Can an artificial intelligence have its own style of music? According to the Professor who developed her, David Cope, apparently, Emily can and does.<\/p>\n

What Electric Dreams Are Made Of<\/h2>\n

In a recent research made public by Google<\/a>, artificial neural networks can<\/em> dream, though they don\u2019t dream of electric sheep. Well, \u201cdream\u201d in this case is used with great freedom, since AIs don\u2019t sleep at all. But how it works is about the same.<\/p>\n

<\/figure>\n
\n
<\/i><\/div>\n
IMAGE: Google Research Blog<\/div>\n
<\/div>\n<\/div>\n

The images above were generated by Google\u2019s artificial neural networks. Basically, the researchers loaded the networks up with images to train it to distinguish one image from another. And in the process, it generates images of what things should look like or at least what it thinks<\/em> they should look like. Talk about Inception<\/em>.<\/p>\n

Practical Uses of AI<\/h2>\n

Obviously, a ton of work is being put into artificial intelligence because it provides a number of invaluable benefits to society.<\/p>\n