Thứ Tư, 3 tháng 8, 2016

Will Artificial Intelligence Make All Business Decisions?

When Google’s AlphaGo computer defeated a human champion in the insanely complex board game “Go” in March, experts were surprised by the computer’s ability to recognize patterns and make quick, instinctive decisions – otherwise known to humans as intuition.

Intuition, or gut instinct, is one of many hallmarks of business decision-making. Steve Jobs had no concrete evidence that the iPhone would be a success, but he saw patterns in the market, and in his mind, that convinced him that consumers would want it.

AlphaGo’s programmers set out to imitate that sense of intuition by first loading 150,000 prior matches between good players into the computer. Then they had AlphaGo play against itself endlessly. As it played, it made billions of self-adjustments to improve its abilities. Eventually, the computer developed a system for quickly judging moves in a manner resembling human intuition.

The same kind of artificial intelligence (AI) system that powers AlphaGo’s human-like intuition is also being used in other areas, such as art and natural language. And there’s even a chance that businesses may take advantage of AI-style intuition at some point. In fact, experts surveyed by Oxford University predict a 50-50 chance that AI will achieve human-like intelligence as early as 2040.

Bias Hurts Judgment 

Bringing AlphaGo’s style of intuitive thinking into the conference room could dramatically improve the way business decisions are made.

Today, companies leave the most important strategic decisions to executives with years of experience and trusted instincts. Yet, lurking beneath those instincts are dozens of unconscious behavioral biases that can lead to deeply flawed decisions. For example, people have a tendency to discount information that doesn’t agree with their assumptions, which turned out to be one of the main culprits in the Great Recession. Or they may choose to pursue short-term gains while discounting long-term risks. And every company has its share of projects that limp along unsuccessfully because leaders decide they have already invested too much to give up on them.

Putting decisions to a group doesn’t eliminate bias, either. Shy people may be drowned out of the discussion. Worse, people may discount an argument based on the personality of the person delivering it.

Leaders Become Scientists

AI systems like AlphaGo could compensate for human flaws. For example, AlphaGo’s intuition is free of bias, and its decisions are backed by massive quantities of analysis. Recognizing and reducing bias and bringing more data into the picture could bring more rationality to the process. Indeed, a Hong Kong venture capital firm already has an AI machine on its board.

Using AI, leaders will become more like scientists. Guided by the overall strategic goals of the company, they will develop hypotheses and hand them over to the AI machines to churn through the data in search of answers. In some cases, AI machines will do the analysis and make the decisions independently, similar to how trading systems work on Wall Street today. Other times, these machines will give leaders the data-backed analysis they need to make the best use of their creative instincts and intuition.

The partnership between decision-makers and AI is already forming. In a survey by consulting firm PwC, executives said that analysis used to inform their next strategic decision will consist of 59% human judgment and 41% machine algorithms.

That balance will likely shift over time if AI machines, like AlphaGo, reach their full potential. Businesses could reach the point where AI makes most of the day-to-day decisions. But creative inspiration will always be needed to fuel new ideas and strategies. It’s difficult to imagine that coming from a machine anytime soon.

For more insight on AI and future tech, see What Artificial Intelligence Can Tell Us About The Future Of Business.

Christopher Koch is the Editorial Director of the SAP Center for Business Insight.

This story originally appeared on The Digitalist Magazine as part of the 10 Weeks of Live Business series.

via SAP News Center

Không có nhận xét nào:

Đăng nhận xét