Technology will cause a shift in the workforce and allow us to focus on higher value tasks such as creative thinking, strategy, and problem solving

There was once an Indian king who loved to play chess. One day, the king challenged a saint to a match, and offered to give her any reward if she won. The saint had an unusual request — she asked for a grain of rice on the first chess square, and the rice was to be doubled on every next square.

The king lost the chess match and being a man of his word, summoned a barrel of rice from the kitchen. He then started placing grains of rice according to the arrangement. 1 … 2 … 4 … 8, and so on. He quickly realised his mistake because by the 20th square he would have to put down 1 million grains of rice. And by the last square, he’d have to put down more than 18 billion x billion grains of rice, which is enough rice to cover India with a meter thick layer of rice.

Imagine that the rice in this story is computing power. According to the law of accelerating returns, our society will reach a point where technology starts expanding so rapidly, that it’ll get completely out of control. Beyond this, it will become impossible to predict the future of humanity — a technological event horizon.

That point has been coined “singularity”

Machines are getting smarter and smarter. This has given rise to a field we’ve all heard about — artificial intelligence (AI).

But what exactly does it mean for a machine to exhibit AI?

First coined in 1956 by computer scientist John McCarthy, AI involves machines performing tasks which are characteristic of human intelligence. This spans across industries and includes activities such as planning, recognising objects, and understanding language. You’re probably interacting with AI more than you think. AI is designed so you don’t realise that a computer is making decisions and this can manifest itself in several different ways. For example, you can speak to Siri or you can text a bot.

Natural Language Processing (NLP) makes these bots more sophisticated so having a two-way communication exchange between machines and humans flows smoothly. Take for example, impress.ai, an AI powered chatbot for recruiters which screens, interviews, engages, and shortlists candidates 24/7. The conversational bot is programmed with NLP which gives it the capability to respond to candidates in real-time.

Also read: impress.ai helps employers screen candidates using AI; raises funding

When AI is enhanced by NLP, it allows technology to adapt to us. Instead of typing out a question, clicking on search, and then browsing through websites for answers, we can simply ask a machine for what we need. In this case, if we’re a candidate being interviewed by impress.ai’s bot and we want to know if the hiring manager has an update, we can ask the bot and receive an instant reply. But how smart are these bots, really? Much like a toddler, machines learn patterns with experience and become more advanced over time.

Machine learning is a way of achieving artificial intelligence

Computer scientist Arthur Samuel coined the term machine learning as, “the ability to learn without being explicitly programmed.” If AI were to be created without the aspect of machine learning, it would need millions and millions of lines of codes with complex rules, parameters, and decision-trees.

Machine learning is a way of “training” an algorithm so it can learn on its own. This “training” involves sharing huge amounts of data to the algorithm and allowing the algorithm to adjust itself and improve.

Also read: Unsupervised machine learning and smart data compression help AI deal with big data

For example, if you want a bot that can recognise a picture of a tree, it’s impossible to explain how to complete this task directly to the bot. You would need to build a bot that “builds” other bots and another bot that “trains” these bots. Then you would need to give the “trainer bot” lots of photos, including photos of trees and an answer key.

The “trainer bot” tests the bots and those that performed best are put to one side, while the others are recycled into new bots. The “builder bot” creates new bots based on the high performing bots with changes, then the bots are tested by “trainer bot” again. The process repeats itself and the best bots are continuously improved. Once the accuracy level is high enough, the machines have “learned” how to identify a tree in a group of photos. Similar to the concept of natural selection, only the best bots will survive.

While it’s exciting to see how fast technology is progressing, it can also seem intimidating to know that machines are able to create more advanced versions of themselves. Hollywood movies like The Terminator and Matrix have popularised the idea that machines will one day outsmart humans and seek revenge on humans.

The more pressing question is: Will AI steal your job?

To put your fear at ease, AI cannot take over your entire job, it can only take over parts of your job, specifically the repetitive and mundane tasks. The AI revolution is just like the industrial revolution. AI, enhanced by machine learning, helps us accomplish more in less time. Technology will cause a shift in the workforce and allow us to focus on higher value tasks such as creative thinking, strategy, and problem solving.

It’s not human versus machine, it’s human and machine versus problem!

—-

e27 publishes relevant guest contributions from the community. Share your honest opinions and expert knowledge by submitting your content here.

Photo by Maarten van den Heuvel on Unsplash

 

The post AI, singularity, and machine learning explained in 5 minutes appeared first on e27.