Managing Cultural and Process Improvements with AI

Seeking Truth from Data, Interesting thoughts.

Managing Cultural and Process Improvements with AI

Published on November 9, 2021. Sam Ransbotham

Professor at Boston College; AI Editor at MIT Sloan Management Review; Host of "Me, Myself, and AI" podcast

For me, this week's biggest news is the publication of our 2021 MIT SMR-BCG artificial intelligence and business strategy research report. While not discounting the substantial financial potential with AI, our research this year focuses on the cultural benefits. Based on a survey of more than 2,000 global managers and dozens of interviews, the report is chock full of examples and data about the cycle between Culture, AI Use, and Effectiveness.

S. Ransbotham, F. Candelon, D. Kiron, B. LaFountain, and S. Khodabandeh, “The Cultural Benefits of Artificial Intelligence in the Enterprise,” MIT Sloan Management Review and Boston Consulting Group, November 2021.

Our key result is that over 75% of global organizations implementing AI report that the technology helped improve their culture. In our fifth year of researching AI and business strategy alongside Boston Consulting Group, we found a wide range of AI-related cultural benefits at both the team and organizational levels. Our report, "The Cultural Benefits of Artificial Intelligence in the Enterprise," outlines these benefits and explains how they relate to financial benefits and competitive advantage.

These financial and cultural benefits do not come automatically to organizations. Instead, they depend on active preparation and ongoing management – two important topics also in recent news.

Developing an Appetite for AI: New Episode of Me, Myself, and AI

Organizations need to have systems and mindsets in place to implement technologies like artificial intelligence, and, often, organizations may not be ready. Sarah Karthigan, AI operations manager for IT at ExxonMobil, joined our Me, Myself, and AI podcast to discuss how she prepares for technology and cultural challenges long before starting AI pilots. She ensures end-users know "under-the-hood" what the tech actually does so that users encourage the necessary changes. Sarah observes that "the partnership goes really, really well once they understand the value that the new solution is able to bring to the table." Active preparation for AI makes a difference and profoundly depends on culture. Plus, if you're curious what Shervin's first time trying sushi has to do with artificial intelligence, this episode's for you.

Managing AI to Promote Financial and Cultural Benefits

Of course, just getting ready for AI is not enough to realize these financial and cultural benefits. Managers are still crucially important. MIS Quarterly, a premier academic journal, just published a special issue on the managerial challenges that come with artificial intelligence. Seven papers address different facets of these challenges.

In "AI on Drugs: Can Artificial Intelligence Accelerate Drug Development? Evidence from a Large-Scale Examination of Bio-Pharma Firms", Bowen Lou and Lynn Wu demonstrate that using AI can accelerate new drug discovery... sometimes, but not always. Innovation depends not only on employees' domain expertise, not just AI skills.

Machine learning tools reduce the costs of repetitive tasks but can introduce systematic unfairness into organizational processes. In Failures of Fairness in Automation Require a Deeper Understanding of Human–ML Augmentation, Mike H. M. Teodorescu, Lily Morse, Yazeed Awwad, and Gerald C. Kane introduce a typology of augmentation for fairness consisting of four quadrants: reactive oversight, proactive oversight, informed reliance, and supervised reliance.

Sarah Lebovitz, Natalia Levina, and Hila Lifshitz-Assaf question the seemingly objective labels organizations use to train AI tools.  Is AI Ground Truth Really True? The Dangers of Training and Evaluating AI Tools Based on Experts' Know-What describes how experts address uncertainty by drawing on rich know-how practices that many ML-based tools do not incorporate.

In Will Humans-in-the-Loop Become Borgs? Merits and Pitfalls of Working with AI, Andreas Fügener, Jörn Grahl, Alok Gupta, and Wolfgang Ketter raise concerns about the loss of unique human knowledge in a host of human-AI decision environments.

Algorithms may produce insights superior to experts by discovering the "truth" from data. But how can systems produce knowledge independent of domain experts yet remain relevant to the domain? When the Machine Meets the Expert: An Ethnography of Developing AI for Hiring (by Elmira van den Broek, Anastasia Sergeeva, and Marleen Huysman) describe how developers navigate this tension when building an ML system to support hiring job candidates at a large international organization. 

Coordinating Human and Machine Learning for Effective Organizational Learning (Timo Sturm, Jin P. Gerlach, Luisa Pumplun, Neda Mesbah, Felix Peters, Christoph Tauchert, Ning Nan, and Peter Buxmann) recognizes that humans are no longer the only ones contributing to an organization's stock of knowledge. 

And finally, Strategic Directions for AI: The Role of CIOs and Boards of Directors (Jingyu Li, Mengxiang Li, Xinchen Wang, and Jason Bennett Thatcher) finds that the presence of a CIO positively influences AI orientation discusses how to build top management teams and boards capable of effectively developing AI orientations.  .... '