AI (artificial intelligence) in business is becoming a ubiquitous phenomenon. AI can potentially transform the operational practice of a business. Efficiencies do exist but they come with a cost and a variety of factors to consider. AI can monitor when human error was made but it can miss its own errors. AI can perform what humans do regularly, such as real time account balancing and inventory resupply reassessments.
As in the medical field, AI has drastically reduced the number of medical errors, as well as provide a host of other benefits. In some areas, AI systems are already capable of matching, or even exceeding, the performances of experienced clinicians. In an interview for Smart Planet, MIT scientist Andrew McAfee, co-author of The Second Machine Age, is convinced, “If it’s not already the world’s best diagnostician, it will be soon.”
AI, if appropriate implemented, can read a business plan and determine vulnerabilities based on variables that it considers. The variables are programed. Algorithm sets the parameters to follow. When it comes to daily functions, AI adds the benefits derived from faster assessments, long term reduced costs, reduced labor costs, enhance customer service, improve feedback to customers, and make customer services and customer account maintenance more personal. It is hard to believe that this is true.
However, as AI presents certain downsides to medial field, in business, the liability remains of the business if something goes wrong. Just as there could be a misdiagnosis in a medical facility, there can be an error in quantifying inventory or assessing risk of an investment of capital, or a realignment of a structure in an architectural plan. for instance, would likely be the responsibility of the presiding physician. In a medical scenario where a faulty AI medical device that harms a patient it would be very likely that the manufacturer, operator, and or the programmer will be held responsible. Indemnities will be part of business plans and arrangements.
As doctors are beginning to learn how to rely on the assessments of AI, businesspeople will experience the same learning curve. Medical personnel remain responsible for errors that may occur. For business this is expected. Business will have to dedicate time and money to assess the reliability or usefulness of information derived from AI, and whether they can have a meaningful understanding of the consequences of those actions and decisions taken.
This inability arises from the opacity of these systems, which—as a side effect of how machine-learning algorithms work—operate as black boxes. It is impossible to understand why an AI has made the decision it has, merely that it has done so based upon the information it’s been fed. Even if it were possible for a business officer to inspect the process, many AI algorithms are unavailable for review, as they are treated as protected proprietary information. Further still, the data used to train the algorithms is often similarly protected or otherwise publicly unavailable for privacy reasons. This will likely be complicated further as business operations personnel grow accustomed to relying on AI more and more and it becomes less common to challenge an algorithm’s result.