10% of Organizations Achieve Significant Financial Benefits With AI — Those That Do Deploy Multiple Human-Machine Learning Approaches, New Study From BCG Finds

Significant financial benefits with AI are noticeably amplified when organizations develop multiple ways for humans and AI to learn together.
Despite increased investment and activity, only 10% of organizations are achieving significant financial benefits with artificial intelligence.
Those companies that draw on multiple types of interaction and feedback between humans and AI are 6X times more likely to amplify their success with AI.

Dr. Sam Ransbotham, Carroll School of Business at Boston College

According to a major study released today by MIT Sloan Management Review (MIT SMR), BCG GAMMA, and BCG Henderson Institute, despite increased investment and activity, only 10% of organizations are achieving significant financial benefits with artificial intelligence. The study highlights the often underestimated role of mutual learning between humans and machines in generating value from AI. Those companies that draw on multiple types of interaction and feedback between humans and AI are 6X times more likely to amplify their success with AI.

The study, as reported in “Expanding AI’s Impact With Organizational Learning,” is based on a survey of more than 3,000 managers in 29 industries in 112 countries and several in-depth interviews with leading experts. It includes a four-year longitudinal examination of cross-industry AI adoption and a variety of use cases. The authors’ analysis found that multiple foundational steps and process improvements enable companies to generate value with AI, but ultimately, companies achieve the most value when mutual learning occurs between humans and machines.

The study also highlights the following investments organizations make to maximize value:

Building foundational capabilities — AI infrastructure, talent, and strategy — increases the likelihood of achieving significant benefits by 19%.
Scaling AI across different use cases and going beyond automation increases likelihood by another 18%.
Achieving organizational learning with AI (drawing on multiple interaction modes between humans and machines) and building feedback loops between human and AI increases that likelihood by another 34%.

Organizations that learn with AI share three essential characteristics:

They facilitate systematic and continuous learning between humans and machines.
They develop multiple ways for humans and machines to interact.
They change to learn, and learn to change.

Organizations that systematically invest in these activities are 73% more likely to achieve significant impact with AI.

“Isolated AI applications can be powerful. But we find that organizations leading with AI haven't changed processes to use AI. Instead, they've learned with AI how to change processes. The key isn't teaching the machines. Or even learning from the machines. The key is learning with the machines — systematically and continuously,” says report coauthor Sam Ransbotham. Organizational learning with AI demands, builds on, and leads to significant organizational change. Additional study data reveals that as of 2020:

David Kiron, MIT Sloan Management Review

70% of global executive respondents understand how AI can generate business value, an increase from 57% in 2017.
59% of global executive respondents have an AI strategy, an increase from 39% in 2017.
57% of global executive respondents affirm that their companies are piloting or deploying AI, an increase from 46% in 2017.
A growing number of companies recognize a business imperative to improve their AI competencies and data infrastructures.
Despite these trends, just 1 in 10 companies generates significant financial benefits with AI.

“The single most critical driver of value from AI is not algorithms, nor technology — it is the human in the equation,” says report coauthor Shervin Khodabandeh. “We continue to see that despite more companies investing in AI technologies and launching AI initiatives, only a small fraction get meaningful value. What this select group do well is that they create integrated AI-Human systems, where AI learns from human and human learns from AI. And the more different ways of learning between the two, the more value there is to get.”

The report features case studies resulting from interviews with senior leaders from companies ranging from retail to energy, legacy to born-digital, across the world.

“Getting significant financial benefits with AI is not the prerogative of digital native companies only,” notes report coauthor François Candelon. “Throughout this research, it clearly appears that success is not bound by legacy, industry, or geography. An incumbent, be it a European energy company like Repsol, an Indian telco like Airtel Bharti or a U.S. retailer like Walmart, can win by taking the right bold moves and make organizational learning with AI become a reality.”

“One big takeaway from this research is that companies need to calibrate their investments in technology, people, and learning processes," adds report coauthor David Kiron. "Financial investments in technology and people are important, but investing social capital in learning is critical to creating significant value with AI.”

Along with the report, MIT SMR and Boston Consulting Group have launched an executive-interview podcast series, “Me, Myself, & AI,” where report coauthors Ransbotham and Khodabandeh talk to leaders successfully leveraging AI in their companies and learn how they did it. The first two episodes, featuring Walmart’s vice president of machine learning and Humana’s senior vice president of digital health and analytics are available here and on all major podcast platforms.

To view more information, please visit www.BCG.com.