Artificial Intelligence fondly called as “AI”, is the talk of the world right now in any business. Even the not so tech-savvy incumbents are trying to change their ideology towards using AI. The reason is because of the fact, that AI helps in transforming their business in to whole new level with reduced prediction cost.
How so? Let us see some basic economics behind usage of AI and it cost of prediction.
What Artificial Intelligence does for any business or its customer is just to Predict. It predicts numerous items in which business and its customers are helped to take informed & rational (pun-intended) decisions.
Prediction & its Costs:
We human predict the outcomes based on our experience, past data, market, environment, & economic conditions. Most of the times the predictions are challenged by any sampling, factoring & biasing errors in estimating.
For example, in demand forecast prediction, when human predicts demand for a high volatile market, the calculation gets complex to estimate. Failure of such prediction results in high prediction cost and brings down the efficiency. Like this, there are many such predicting activities in the value chain of any business.
Use of Artificial Intelligence also starts with failure with high prediction cost. But it learns from the failure and provides better predictions without any human biases. It starts in incremental phase of prediction which are applied in every stage of value chain to bring in efficiencies.
For example, Prediction is used in inventory management in estimating the inventory turnover, or how much to order or produce, in traditional assembly lines to predict the productivity, in supply chain to predict the supply requirement, in autonomous vehicles to move around in restricted workspace.
Soon when AI gains experience in predicting the traditional problems, it can scale up the prediction to more complex problem such as autonomous driving in city, predicting customer needs and desires. This recast of prediction results in significant reduction of prediction costs. This low-cost can be harnessed to exponentially scale the AI prediction process and transform the business model.
Human Involvement & AI in Business Model Transformation:
The substitute of AI prediction is human prediction. When the value of AI prediction increases, the value of human prediction decreases. This does not mean that humans will lose their job to AI. Because, increase in AI prediction value results in increase in value of AI complements such as data, judgement & decisions and actions.
Humans are still needed to make judgement based on the AI prediction. Humans decide the course of action which needs to be implemented. During the course of AI prediction, Data become a high value item, in which humans are needed to mine, & cleanse, monitor, understand and measure. Once this start creating value, the whole business model will go for major digital transformation.
For example, AI prediction is used in Amazon recommends a list of products (from its millions of products catalogue) to its customer based on his/her interests. This make customer to stick with Amazon and makes them buy more products from them.
As more customers buys more products from amazon, the huge amount of data gathered helps AI to fine-tune its predictions and scale the sale. This transforms the business model from traditional reaching out to customer to customers coming to amazon. As this AI prediction accuracy scales exponentially, the volume of business increases as well as the profits.
This Blog is inspired from McKinsey Quarterly interview article “The Economics of Artificial Intelligence” by Rotman School of Management professor Ajay Agrawal. Professor Ajay Agrawal is also author of the book, “Prediction Machines: The Simple Economics of Artificial Intelligence”, coauthored with professors Joshua Gans and Avi Goldfarb, Agrawal
Image used in this Blog Courtesy: youthsera, edgenetworks