Solving the manufacturing workforce challenge in the age of agentic AI

AI in supply chain is changing how to capture tacit knowledge, what is done on shop floors and the qualities of leadership.

In brief

• AI agents can automate and codify tacit knowledge, long a stumbling block for manufacturers, with many AI solutions hitting the market in the past 12 months.
• But supply chain management must confront how their jobs will also evolve. Transformation often lives or dies based on culture rather than just technology.
• Companies have little wiggle room on materials costs. But those that tilt the labor equation in a more favorable direction gain a huge edge in the market.

Workforce skills are an often-overlooked leg of the Industry 4.0 stool, alongside digital manufacturing solution deployment and work process improvement. And while many manufacturers have struggled in this area, the advent of agentic artificial intelligence (AI) in supply chain is flipping preconceived notions on their head about what tasks need to be done and how, while raising questions about the skills that managers and frontline leaders, not only blue-collar workers, need most.

For instance, you could film a senior operator doing a task, such as setting a center line on a machine or performing maintenance or calibrations. Analytics platforms can then pick apart how the recorded task is completed, creating a standard operating procedure, a training guide or a step-by-step video, all with the goal of standardizing production management processes as they are actually performed on the shop floor.

Additionally, innovative tools, such as digital twins, augmented and virtual reality, and connected worker applications, can then facilitate the transfer of tacit knowledge. They can act as the technology process backbone that will guide work execution, regardless of whether the work is being done by agents or people. These technologies enable real-time sharing of expertise and create immersive learning experiences that help bridge the knowledge gap.

Agentic AI adds a disruptive element to workforce training in that this tacit knowledge can form the basis of automated processes, not only reducing reliance on individual expertise but also upending the norms of how work is completed. And the implications may cut against the grain of what you’d expect from your workforce.

The crisis of confidence among managers

Senior leadership at manufacturers generally doesn’t need much persuading about the power of AI and digital and the need for upskilling. And while the workers on the shop floor may signal some resistance early on, they are more cognizant of how supply chain technology can make their lives easier, minimizing the time-consuming, transactional work that must be endured. One company that has been transitioning to automated, unattended operations is doing it to improve the employee experience by eliminating the need to staff unpopular night shifts.

Yet the managers in the middle, who comfortably fulfill a niche in today’s ways of working, often feel trepidation — because they, too, need to reskill, and most companies’ workforce modernization platforms aren’t addressing the leadership level. For instance, a big leadership characteristic is motivating people to do their best. What does it mean when rallying the troops to win the day includes using AI agendas that will be fulfilling many tasks on the critical path? Are these managers equipped to interpret data to validate agent decision-making, or is it in an enterprise resource planning (ERP) system they can’t access or don’t know how to use? How do they orchestrate a team of people plus agents?

The EY Agentic AI Workplace Survey from mid-2025 affirms these feelings: Half of managers doubt their ability to lead AI-augmented teams, and most expect management to become harder, not easier. And among manufacturing-specific respondents, 54% of managers say that addressing job security and role changes is critical for successful AI adoption (compared with 43% of manager respondents overall).

Four tactics to deploy today

Putting agentic AI in supply chain management to work will define the market for a generation to come. It should be invigorating to be at the forefront of this technological shift because it also provides an opportunity to reverse the trends in manufacturing — to shed the traditional, conservative reputation of the sector and appeal to a new cohort of digital-native workers previously drawn to other industries with data-driven jobs that are literally building the future. Four tactics to consider include the following:

1. Promote transparency around plans to turn anxiety into eagerness. What is the future vision you’re working toward — is it to do the same work with increasingly fewer people or to achieve more with new capabilities that build on the strengths of both people and technology?
2. Establish continuous learning systems so that employees can keep pace with new developments and evolve their skills. It is essential to adopt a multimodal upskilling model that simultaneously addresses immediate skill development needs and prepares the workforce for future demands. Crucially, this is not just necessary for those on the shop floor but also for those who manage them.
3. Confirm that all workers possess a solid understanding of data-driven decision-making and digital tools, which are critical in a technology-driven environment. When something is off, would they recognize it? And would they have the tools and the know-how for validation?
4. Cultivate leaders who can navigate change. Who is willing to experiment and expose themselves to a thrilling new opportunity amid great uncertainty? Challenge your notions about what a manager even looks like.