Thank you for Subscribing to Energy Business Review Weekly Brief
Artificial intelligence - Algorithms of Change
Ash Dhupar, Chief Analytics Officer, Publishers Clearing House
There is a lot of discussion around artificial intelligence (AI) and how these algorithms can change the pace of mankind’s next evolution in technology. As always, there are two sides to the story. There are those who are afraid and those who see the opportunities that AI will bring to our lives. As a practitioner, I feel there is a need for us to understand whether AI is a science fiction or a part of our everyday lives.
The fantasy, fiction, charm and promise of AI has kept the scientific community to continue pursuing it for last 70 years. AI has been at times hot and at other times cold since its inception in 1950’s. From when Alan Turing ran the first experiment using computers to emulate human intelligence, to the first self-driving cart developed in 1960’s by Stanford University. In the 80’s and 90’s the AI hype slowed down, and to a certain extent was mocked, except for Hollywood through movies like The Terminator, The Matrix and more recently Ex Machina. In the last decade the rise of computing powers has given the much-needed boost to propel AI to a new level. The computing power of the iPhone X has a million times more power than all of computing power NASA had in 1969 when it launched Apollo 11. If a phone in our hands has that much power, you can only imagine what the fastest computers can do, drum roll… a whopping 93 Petaflops. A Petaflop is one thousand trillion (Quadrillion) floating operations per second. It’s this sheer power that is being utilized by scientists to make AI so intelligent and so useful.
For those who would ask, so what is AI? In simple terms, it’s making computers learn the way humans think. In technical terms AI has two fields; machine learning and deep learning. Both fields of AI are affecting our everyday lives today and we will continue to see affects in major ways every day. When we buy products from a retailer’s website it’s very common to see recommended products that we may like. These recommendations which can account for 25 percent to 30 percent of sales for an organization are powered by machine learning algorithms. On the other hand, when we give a voice command to Siri to give us a weather update, it is powered by deep learning algorithms. These deep learning algorithms are behind the much advertised, machine beating humans in Chess, the game, GO and the gigantic work done to date in self-driving cars.
"There is no doubt that AI will change the evolution of technology and reface the marketing and science world"
In my opinion the potential of AI is the next chapter in our technological evolution and will continue to impact our lives in many ways. Areas like Robotics and Internet of Things (IoT) will use the trio power, a combination of computing power, AI algorithms and data to transform the customer experience. The progression of computing power and algorithms are fairly matured with enough companies, research labs and investment dollars behind them. However, the area which is in its infancy stage of evolution is collecting, cleaning and securing data.
Let me give you a peek into the world of IoT to clarify this infancy aspect. IoT is one of the emerging areas to leverage AI and will have a significant impact on data collection and quality. IoT experts who are thinking of eliminating traffic problems in congested cities like New York will have to collect data by installing IoT sensors on every street corner. Imagine the data to be collected and fed into the real time AI algorithms to make the split-second decisions for controlling the flow of cars, bikes, and pedestrians simultaneously, with no room for failure. Currently, majority of organizations are not prepared for ingesting and processing such vast quantities of data. This will trigger a meteoric rise in companies and technologies to address these issues in nearly every industry, from agriculture to advertising. To err is human, but to not err is Robotics. Robotic Process Automation (RPA) projects, which are designed to reduce error and increase efficiency, are making extensive use of AI algorithms. The algorithms primarily use unsupervised machine learning to rip through massive amounts of data and classify them in quality buckets. These robotic automations are at infancy stage of bringing quality data processing to be more mainstream.
As a CIO, CDO or CAO are we ready to take advantage of AI to enhance the customer experience and shareholder value for our organizations? The AI based customer experience is not a fantasy or thing of the future. In fact, progressive companies are heavily invested in making momentous impact on their customers using AI. These companies are transitioning their marketing and advertising from traditional segmentations strategies to micro segments or in some cases to a segment of one. The road to adopting AI based decisioning can be complicated or simple, it all depends on our approach. If you are just starting, my advice is don’t try to achieve the Holy Grail solution on day one. Start simple, the use of machine learning algorithms doesn’t require heavy investments and it can show immediate impact to your organization’s bottom-line. The full power of AI does require investment in talent, technology, and process, which is not easy to accomplish without a commitment from your board and senior management.
Organizations are laying AI foundations with the aspirations of making a positive, noteworthy difference with their customer’s experience. Certainly, AI is not taking over humanity as depicted in the science fiction movies and for the foreseeable future we are in control to unleash the power of AI to reduce imperfections, increase efficiencies, and improve shareholder value for our companies. Yes, change is the only constant, but the rate and pace of change with AI will be unparalleled to what we have seen in the last 50 years. There is no doubt that AI will change the evolution of technology and reface the marketing and science world.
This content is copyright protected
However, if you would like to share the information in this article, you may use the link below: