Can Artificial Intelligence (#AI) replace my job?

One of the most popular and dreaded question being asked in the industry today, is that whether AI can replace my Job. 

Actually, it depends in how we (as a service professional) choose our battle against AI. If you compete with AI on non-interesting, repetitive, non-creative , highly multi tasking, tedious, large data oriented and long running tasks, we have little chance to win, and soon our job will be replaced by Intelligence bots.

In order to understand it, let us go back in history.

AI has been around for 60+ years now. During 1957, when Rosenbatt discovered Perceptron, closely mirroring neuron in human brain, it was speculated that AI will replace every human job. At that time, The New York Times reported the perceptron to be "the embryo of an electronic computer, and it will be able to walk, talk, see, write, reproduce itself and be conscious of its existence." Millions of dollar were invested in it by IBM to make it a reality. However, few years later, Minsky (another scientist) published a book called "Perceptrons" demonstrating the limitations of a Perceptron (precisely single layer feed forward neural network). All hopes doomed, and suddenly all the funding stopped on AI. The Dead End. Though, AI and neural networks got attention again after a decade, with a new energy and enthusiasm.

In 1997, an IBM computer called Deep Blue defeated world champion, Gary Kasparov, in the game of chess. However, no AI program could defeat humans on game of GO since 2016. The mean reason is that GO requires complex and involved visual based thinking pattern, on the other hand, a chess is purely a search problem in a large tree of possibilities. Similarly, in 2016, IBM developed Watson could crawl Wikipedia and huge collection of documents to store and index the information in such a way that it could defeat two greatest human players in a pure language game called Jeopardy!!.

The main crux of the story is that AI, rather all computers, has been good in two types of problems:

1. Known Problems and Solutions, where the algorithms are already known such as multiplication of very large numbers or playing a chess game, indexing and searching documents, optimization problems etc.

2. Finding patterns in a huge data such as finding genes in human DNA, reading whole Wikipedia and answer questions.

At the same time, there are many frontiers where AI is still struggling to catch up

1. Visual Objects recognition.
2. Natural Language Processing and Understanding.
3. Voice recognition and understanding.
4. Interacting with real world objects.
5. Recognizing facial emotions. 

Note, many of the above tasks are done effortlessly by a 5 year old kid with a great precision.

There are also many frontiers that AI research has not even started or is in very nascent stage

1. Consciousness  
2. Any Creative activity such as Music, Painting, Writing
3. Given a PRD, write a software.
4. Gut feelings
5. Laughing, Crying , Happiness
6. Learning new concepts
7. Define Problems
8. Many more.

If you notice a pattern, the current area of AI research is focusing on our 5 senses, body movements and applications related to unifying them together such as robotics and driver-less cars, and with abilities that computers are already good at such as storing, indexing and retrieving information, multi tasking, optimization etc.

The tasks that involve area of the brain, developed later during the human evolution and deal with complex concepts such as learning, emotions, intuition, creativity, love and spiritual, are still beyond the reach of AI.

So, great coders, hackers, architects, algorithm designers, data scientist, artists, sales person, strategic thinkers, innovative people, automation folks, good managers and anyone who keeps improving his/her skills need not to worry. 

Finally, these are my personal opinions based on my understanding of AI. If you think otherwise or you have additional points to add, feel free to comment below.