How companies should — and should not — deploy AI
https://arab.news/we9hf
Even though nearly half of office workers now turn to generative AI in their daily work, fewer than one in four CEOs report that the technology has delivered its promised value at scale. What is going on?
The answer may lie in the fact that generative AI was initially presented as a productivity tool, which led to it being strongly associated with cost-cutting and workforce reductions. Spotting the risk, some 42 percent of employees surveyed in 2024 worried that their job might not exist in the next decade.
In the absence of training and upskilling to harness the technology’s potential, it is not surprising that there would be more resistance than enthusiasm. Like antibodies fighting off a foreign body, there can be an “immune response” within organizations, with employees and managers alike resisting change and looking for reasons why AI “won’t work” for them.
In addition to slowing adoption, such resistance has prevented a fuller exploration of other potential benefits, such as improved decision-making, enhanced creativity, the elimination of routine tasks and higher job satisfaction. As a result, there has been little consideration of how to “reinvest” the time that AI can save.
AI adoption is not about saving minutes. It is about reinventing work for the benefit of employees and the organization.
Vinciane Beauchene and Allison Bailey
Yet our research finds that employees who use generative AI regularly can already save five hours per working week, allowing them to pursue new tasks, further experiment with the technology, collaborate in new ways with coworkers or simply finish earlier. The challenge for business leaders, then, is to emphasize these potential benefits and provide guidance on where to refocus one’s time to maximize value creation.
Consider the example of a global healthcare provider that recently deployed generative AI across its 100,000 employees. It created a scalable AI learning program with three objectives: high AI literacy across the organization, so that all employees could make the most of the technology; a broad suite of AI tools for every work scenario; and compliant usage. Owing to this holistic approach, the company soon improved employee satisfaction and productivity at the same time.
But AI adoption is not about saving minutes. It is about reinventing work for the benefit of employees and the organization. When a company treats generative AI merely as a timesaving tool, it is more likely to chase piecemeal use cases — 10 minutes saved here, 30 minutes saved there — which will not have a meaningful impact on the overall business. After all, small-scale AI applications that yield diffuse productivity gains are difficult to reinvest or capture on a profit and loss statement.
Without a holistic strategy to redesign their core processes around AI, organizations risk optimizing isolated tasks rather than fundamentally improving how work gets done. The result, all too often, is that bottlenecks will simply be relocated to other parts of the process or value chain, limiting overall productivity gains. For example, in software development, an AI that speeds up coding can lead to more arduous debugging or other delays, negating any efficiency gains. Real value comes from integrating AI across the entire development lifecycle.
This example also raises a larger issue: Too many organizations pursue scale without first reimagining the structures and workflows needed to harness cumulative gains. The usual result is a missed opportunity, because time savings that are not reinvested strategically tend to dissipate. Rather than adopting a let-a-hundred-flowers-bloom approach, organizations should pursue a few big transformational initiatives focused on reimagining work from end to end.
The true promise of generative AI lies in unlocking what we call the “golden triangle” of value: productivity, quality and engagement/joy. An AI strategy should reimagine workflows to eliminate inefficiencies, augment decision-making and processes to encourage innovation and creativity, and enhance work, not mechanize it. Employees are more likely to embrace AI enthusiastically when it eliminates drudgery, feeds creativity and accelerates learning. Proper attention to upskilling will ensure that the technology augments human potential, boosting workplace engagement and job satisfaction.
By emphasizing engagement and the quality of experience alongside productivity, organizations can move beyond a cost-driven perspective to one that creates more value for the business, its employees and its customers. AI can be much more than an automation mechanism, provided that firms adopt a comprehensive strategy for deploying it.
Organizations should pursue a few big transformational initiatives focused on reimagining work from end to end.
Vinciane Beauchene and Allison Bailey
Business leaders should keep five imperatives in mind. The first is to focus on the biggest pools of value with the best-defined business cases for integrating AI. The second is to reimagine work, rather than simply optimizing it. AI should be used to transform entire workflows, not just automate a few steps.
Third, managers must invest in upskilling, so that everyone understands the technology and its potential. Fourth, the golden triangle, with its balance between productivity, quality and employee engagement/joy, should be businesses’ golden rule.
Lastly, organizations should measure value beyond cost savings. Businesses that deploy generative AI most effectively will track its effects on workforce empowerment, agility and new revenue streams, not just operational costs.
By heeding these imperatives, companies can use AI as a force for reinvention, rather than just a productivity tool. In the process, they will set the pace for the next era of business.
- Vinciane Beauchene is Managing Director and Partner at BCG, where she serves as Global Lead on Human x AI.
- Allison Bailey is a senior partner and Managing Director at BCG, where she serves as Global Vice Chair for People and Organization Practice. Copyright: Project Syndicate