Artificial Intelligence—which we define as information systems and applications that can sense, comprehend and act—has moved beyond the research lab and captured the attention of CEOs and other high-level executives.
The media regularly draws attention to innovative business solutions based on AI. Venture capitalists are funding AI start-ups at a rapid pace. Technology companies are moving swiftly to create and capture value in this emerging area.
Decision makers must recognize that AI isn’t a matter of any single technology or application—whether driverless cars, smartphone virtual assistants or myriad of other examples.
In this report, we offer a perspective on how these technologies affect business and society, and present a framework for understanding how AI can deliver value for your organization and industry.
AI systems can be self-learning, like bright students who are given educational materials and then can learn by themselves. That is, computers are enabled to do the following:
Consider how a border-control kiosk uses computer vision technologies such as facial recognition to sense characteristics of travelers.
Integrated with other technologies such as multispectral image analysis (scanning passports using infrared and ultraviolet light), extensive information databases and matching algorithms, an integrated solution here can improve security by identifying people on unauthorized entry lists or others posing a risk.
AI systems also comprehend through technologies such as natural language processing, inference engines and expert systems.
These technologies have a wide range of applications across multiple industries. For example, a medical diagnostic system can help doctors identify diseases and suggest treatments.
An AI system acts independently, taking action within a process through technologies such as inference engines and expert systems, or it can direct action in the physical world.
Consider the driverless car, which senses the environment, understands the myriad inputs and then steers the car without assistance from a human driver.
The range of available AI solutions can be categorized by the complexity of the work that is being done, and second, the complexity of the data and information being worked with. The range of solutions can be classified into four primary types of activity models.
The efficiency model characterizes, which characterizes more routine activities based on well-defined rules, procedures and criteria. The goal here is to provide consistent, low-cost performance.
In this model, work is more likely to involve judgment and is highly reliant on individual expertise and experience—activities performed, for example, by doctors, lawyers, financial advisors and engineers.
Decision-making and action is generally taken by humans themselves, while technology’s role is to augment human sensing and decision making—enabling analysis and offering advice and implementation support.
The goal is to improve the overall ability of workers and companies to produce a particular desired result. This class of workers typically requires considerable knowledge of their industry, company and business processes.
Their success is highly reliant on coordination and communication and involves a wide range of interconnected activities—work such as administration, managing, sales and so forth. In these solutions, technology acts as personal assistant or agent on behalf of humans at their direction.
In this model, AI solutions enhance creativity and ideation by humans—activities and roles such as biomedical researchers, fashion designers, chefs, musicians and entrepreneurs. Humans make decisions and act, while technology helps identify alternatives and optimize recommendations.
Game-changer for every industry
A key is not to become too entranced by any particular technology, as if that technology by itself is the answer.
It is vital to think first in terms of types of work, and then consider the business rationale for integrating technologies into a total AI solution related to that work.
It is important that business continues to engage in the ongoing dialogue about these technologies’ effects on jobs, education and society.
What happens in terms of the social impact of AI is not up to the technology, but to us. The business opportunity of getting it right is too significant to be left to chance.