Artificial Intelligence – A Shift From Traditional to Intelligent Experiences

By: Ashwin Ramachandra, VP and Head, Product Engineering Services, Sasken Technologies 

The concept of artificial intelligence has been around for decades and has considerably improved. The term AI was coined by John McCarthy in 1956 with a goal to develop machines that are intelligent and are useful in the real world scenario. But ever since IBM’s Deep Blue defeated the chess wizard, Garry Kasparov in 1997, AI has played a much more genuine and beneficial role in technology.  A lot has changed from then to now. Technology is much more easily available and we see people skillfully applying different technologies, using toolkits and frameworks, to a wide variety of problems in the industry from design to quality control to manufacturing to customer service.

The thought process employed to arrive at certain deductions and conclusions have been kept in intangible form for centuries. Because of this, the initial human capability to alter surroundings had been slow and steady until the last hundred years. But, a change in man’s resolve to somehow imprint those steps into the tangible world has led to the creations of tools. Subsequently, tools lead to machines, machines paved the way for instruments to look into the microcosm and the macrocosm. And finally, the insights from these worlds led to the revolution witnessed by mankind over the last five decades – The Computing Revolution!

Now your GPS is smarter than you are in spatial navigation, and Google is smarter than you are in long-term memory. Apple’s Siri and Amazon’s Echo are some familiar AI examples for you to relate. However, it wouldn’t be wrong to say that we are merely in the first hour of the AI revolution. Also, AI alone will never emerge as an industry inclination as it is not a single type of technology. Artificial Intelligence paired with Machine Learning is predicted to be the most sought after technology trend in the near future. Machine learning refers to algorithms that enable software to improve its performance over time as it obtains more data. This is programming by input-output examples rather than just coding.

IBM Watson, a cognitive system designed to understand data and reason accordingly, leverages natural language processing through cognitive APIs. The system uses machine learning to analyze data and learn from it, to eventually derive insights and scale over a period of time.  “There is value to be gained from systems that go beyond general abstractions and reason in specialized ways,” believes IBM.

Advances in AI are based on processing, algorithms, and a lot of data. Graphical Processor Units (GPUs), capable of conducting linear algebra, have been a breakthrough in the AI space. Going further, advanced specialized chips will be used to leverage deep learning for self-training machines.

Speaking of data, we have more data available today than ever before. However, it is exciting to know that companies readily open source their machine learning algorithms but they never give you their data. Amazon will never share their buying behavior data and Google will never share their search data. Therefore, it suggests that data triumphs the algorithms. For us to have an impact on AI, one of the biggest opportunities is Internet of Things since IoT is going to provide us a lot of real world data about who we are and what we do.

As reported by Forbes early this year, “IDC estimated that the AI market will grow from $8 bn in 2016 to more than $47 billion in 2020.”  While another report suggests, “Business spending on MI is forecast to reach $31.3 billion by 2019, according to IDC.” To quote Sundar Pichai, CEO of Google, “The last 10 years have been about building a world that is mobile-first. In the next 10 years, we will shift to a world that is AI-first.”

With such intense expectations and growth predictions, AI is definitely not just a fad! Let us look into some upcoming AI trends in industry applications.

• Improved Healthcare: AI will drive a forthcoming revolution in healthcare, fuelled by new classes of data – genomics data, wearables data, and real-time data of the human organism under certain conditions. Healthcare givers will be able to combine unparalleled pattern recognition with other peoples’ data through crowdsourcing. Consequently, they will be able to separate out a particular mutation and its expression to see how it shows up in the population in every individual who has that mutation.
• Autonomous cars: AI will catalyze the advent of robust self-driving cars. Although we have them in use on small pilot tests all over the world, we will see them coming in force and dramatically decrease the casualties on the highways due to distracted driving.
• Intelligent apps: Tried buying furniture from Amazon lately? Wonder why the app suggests a specific set of furniture to buy the next time you login? This is because developers are now incorporating AI in the apps to enhance user engagement. These apps identify user behavior to provide smart responses, thereby self-learning and evolving over a period of time. That is exactly how Uber is using AI for route optimization. A similar logic goes behind Netflix using AI to go from recommendations based on what you have seen, to what you like.
• Better hardware: The inability of the hardware to support intelligent systems was inhibiting the spread of AI. Moreover, the rise of intelligent systems would not only demand an improvement in the software but also in the hardware. To simplify user interactions with the machines, precise sensors would be integrated. Google’s deep learning mimics the structure of the human brain by using neural networks but doesn’t follow its functions exactly. Going forward, GPUs would upgrade AI systems by expediting the Deep Neural Networks.
• The combined power of AI and Cloud: Staying in line with the current trends require businesses to move to the Cloud. With big players talking about combining the capacities of AI, machine learning and aided technologies with the cloud, we can expect more AI systems in the enterprise scenario than ever.

Thinking different is the engine of creation, wealth, and a new economy. It is this approach that led humans to hit the magical moment when they realized that they can program intelligence. Nevertheless, the manifestation of AI and how it is going to transform the society depends on how companies chase the next new idea. Until then, we know the question of the hour is not ‘if’ but ‘when’ AI will have a huge impact on our industry.