Recent advances in artificial intelligence (AI) and cognitive computing are delivering algorithms that outperform humans in many types of analysis and that are capable of making decisions and taking action unaided by mere people. Barry Devlin - a foremost authority on business intelligence (BI), big data and beyond - explores what it might mean for BI when self-service is displaced by bot-service.
The progression in BI and analytics through descriptive, operational, diagnostic, predictive to prescriptive is often used as an indication of organizational maturity in decision making support. With digitalization of business ongoing, the advice is generally that we must aspire to the prescriptive level. Understanding what that prescriptive level would look like means thinking about algorithmic decision making and current advances in artificial intelligence (AI).
Artificial intelligence is impacting all areas of work and computing, and decision making will not be immune. Like these other areas of life, BI will be speeded up and/or improved, so the drive for business value will make AI increasingly attractive in BI. AI (in simpler forms) is already widespread in operational BI / decision management, where algorithm use has already been growing for years to automate simpler decisions like routine credit offers, airline ticket pricing, or up-sell/cross-sell offers on websites. As AI gets cleverer, more tasks can be automated and more complex decisions can be handled. We can thus see automation creeping up the "decision making value chain" even into some areas of tactical BI.
Augmentation, like automation, is a method of implementing AI. Augmentation is where AI is used to extend the abilities of the human decision maker to better understand big data, to offer statistical advice, and most interestingly to propose useful approaches and solutions to real-world problems (vs. today's BI, which only offers data and visualizations of data). Augmentation and automation use the same underlying data/information to "understand" context, weigh options via neural networks, etc. but while automation actually makes the decision, augmentation defers to the human.
Automation can certainly replace human workers in many tasks and so may be seen as threatening jobs, particularly in more routine decision making. Perhaps more worryingly, in both automation and augmentation, introduces a ‘black box’ into decision making. Often it is impossible to describe how the AI came to a conclusion/recommendation/decision. Some see this as good (eliminate human prejudice), but I see it as threatening, because prejudice may exist in the algorithm - intentionally or not - unknown to its users. This may also contravene EU law - the General Data Protection Regulation, due to come into force in 2018. There are many issues here that have, so far, not really been thought through. We need to do this before we open Pandora's box. If decision-making should be in the hands of algorithms is the key question; and there are opposing views. I believe that the profit motive will tend to push for more automated decision making, while the public good would push in the opposite direction... it will have to be a balance in the end.
In this era when computing power will certainly replace decision making to a certain extent, AI will affect the roles of decision makers. Whether in automation and augmentation, repetitive work will shift to the AI machine. So, managers will need to upskill to more ‘strategic’ and ‘creative’ thinking. They will also need to become knowledgeable in how the algorithms reach conclusions (in as far as this is possible). I suspect that many middle management roles will disappear.
Dr. Barry Devlin is a founder of the data warehousing industry, defining its first architecture in 1985. A foremost authority on business intelligence (BI), big data and beyond, he is respected worldwide as a visionary and thought-leader in the evolving industry. Barry has authored two ground-breaking books: the classic "Data Warehouse-from Architecture to Implementation" and “Business unintelligence -Insight and Innovation Beyond Analytics and Big Data” in 2013.
Would you like to know more about the effects of AI? Visit the keynote speech of Barry Devlin on this matter at 360º BI: Business Intelligence Conference 2017 ' on February 7 , 2017.