The AI advantage in decision-making

We live in a data-centric world, with artificial intelligence (AI) as the engine that increasingly drives decision-making. Companies and government agencies all seek to make smarter decisions, often working with incomplete data, to gain competitive advantages. Whether it be uncovering the root causes of crime rate increases, policies to reduce avoidable gun deaths, early detection of cancer or forecasting weather events, data drives many organizations’ decision processes. 

Data-driven decision-making is the catchphrase of government and industry leaders, and even elected officials who want to sound “tech savvy.” When it comes to data, “follow the science” means “follow the ‘data’ science.” Yet, data of itself is insufficient to make better decisions.

Focusing solely on data is akin to only looking in the rear-view mirror when driving your vehicle on a busy highway. If you do not concentrate on where you are going, but rather, direct your attention only on where you have been, you are likely to cause an accident. 

Why is data so important?

Data contains information. The challenge is extracting such information to inform better decision-making. Therefore, data contains potential that can be unleashed with the appropriate tools.

AI models and algorithms are the workhorses of decision-making. They are designed to systematically extract useful information from data. Such information is then used to provide predictions, forecasts or interpretations that support decision-making. Whether it be self-driving automobiles, disease diagnosis to improve health care or helping you make a better online purchase, AI continues to find new applications that make our lives better, safer and hopefully more fulfilling. 

For most people, these models and algorithms are completely invisible, and often incomprehensible, for good reason. They ae designed using advanced concepts from mathematics and statistics that few are trained to appreciate. Yet, these models and algorithms routinely produce remarkable results, informing decisions that are often far better than a person on their own may be able to muster. This is the value of data, and the power of models and algorithms that use such data to make smarter decisions. 

Who is designing such models? 

Computer scientists are doing much of the heavy lifting. These researchers in academia as well as government and industry labs (like Google Research) are designing models and algorithms that permit data to be used as inputs, with new output data generated to inform smarter decisions.

Models that embody artificial intelligence with esoteric names like machine learning, neural networks and Markov decision processes are making sense of data so that decision-making can be enhanced. These models and algorithms effectively turn rear view mirrors into crystal balls that can look into the future. 

A word of caution.

There are no free lunches when it comes to AI and data-driven smart decision-making. This is why the ethical use of AI demands its own AI Bill of Rights. Without such guard rails, the untethered application of AI may erode the very liberties granted to us all by the Bill of Rights.

AI models are the tools that facilitate smart decision-making. This makes computer scientists the unsung heroes on the decision-making playing field. Decisions would be less certain without such models and algorithms. Yet, with such tools, results are attainable that just a few short years ago were inconceivable. This explains why industrialized countries, including China, Canada and Russia, are investing so heavily in AI to gain economic and security competitive advantages.

Indeed, competitive advantages gained by AI may not just feed more people, improve transportation or enhance health care, it may also be needed to win the next war or explore space.

AI is at the fulcrum of the current technological revolution. Data may be the fuel for this revolution, but AI models and algorithms are the engine that is propelling it forward.

Sheldon H. Jacobson, Ph.D., is a professor in computer science at the University of Illinois Urbana-Champaign. A data scientist and operations researcher, he applies his expertise in data-driven risk-based decision-making to evaluate and inform public policy.