Middle Management in the Age of AI: A Crisis Nobody Is Planning For

As AI takes over routine tasks below and analytical work above, the layer of leadership holding most organisations together is quietly becoming redundant

Video summary: https://youtu.be/twceZ-bBcr4

Sign up for my Substack daily AI newsletter here.

See my AI Training course portfolio for corporate Business Leaders here.

Article link: https://open.substack.com/pub/johanosteyn/p/middle-management-in-the-age-of-ai?r=73gqa&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true

Follow me on LinkedIn: https://www.linkedin.com/in/johanosteyn/

In almost every organisation I work with, the conversation about AI follows a familiar pattern. Leaders talk about automating frontline processes. Boards discuss AI-driven dashboards and executive decision support. Pilot projects target customer service, compliance, and back-office efficiency. What almost nobody talks about is the layer of people sitting between the frontline and the boardroom: the middle managers who coordinate, translate, escalate, and absorb pressure across the organisation every single day. 

AI is now moving into their territory from both directions at once. The crisis this creates is not theoretical. It is already unfolding inside organisations that have no framework for addressing it and no honest conversation about what it means for the people caught in the middle.

CONTEXT AND BACKGROUND

Middle management has always occupied an uncomfortable position. Too senior to be shielded from organisational politics, too junior to set the agenda, middle managers have long been the connective tissue of large organisations — translating strategy into action, managing performance, resolving conflict, and keeping information flowing between layers that rarely speak directly to each other.

For decades, this role was secure precisely because it was complex and relational. Machines could handle repetitive tasks, but the judgment, context-reading, and human coordination that middle managers provided seemed beyond automation. That assumption is now being tested. According to the World Economic Forum’s Future of Jobs Report, published in 2025, nearly 40 per cent of core work skills are expected to change within the next five years, with AI and automation driving the largest share of that disruption. Middle management roles are among those most exposed to restructuring as organisations redesign around intelligent tools.

The pressure is coming from two directions simultaneously. From below, AI agents are taking over the coordination, scheduling, reporting, and performance monitoring tasks that middle managers have traditionally owned. From above, AI analytics tools are giving executives direct access to operational data, workforce metrics, and strategic scenarios that previously required a layer of human interpretation and synthesis. The middle is being compressed.

INSIGHT AND ANALYSIS

A recent episode of the Valuetainment podcast hosted by Patrick Bet-David illustrated this vividly. Panel members described AI agents handling morning briefings, news analysis, sales reporting, and email prioritisation — tasks that in many organisations would previously have involved a manager reviewing, summarising, and presenting information upward. When a single professional can automate 15 tasks daily and compress a full day’s work into two hours, the question of what the layer above that person is still doing becomes harder to avoid.

This is not simply a story about efficiency. It is a story about organisational design. McKinsey's research into the agentic organisation describes a future where small teams of two to five people supervise agent factories of 50 to 100 AI agents running end-to-end business processes. The report confirms that this is already happening: a healthcare company replaced a traditional ten-person software development team with a three-person unit overseeing AI agents. In practical terms, this means fewer managers overseeing more processes — and those managers needing fundamentally different skills than the ones that got them promoted in the first place.

The human dimension of this shift is being dangerously underestimated. Middle managers are not an abstract category on an organisational chart. They are experienced professionals who have built careers, developed expertise, and made significant personal investments in their roles. Many are parents. Many are the primary earners in their households. When organisations restructure around AI without honest communication and genuine transition support, the cost falls on real people in ways that do not appear in productivity dashboards.

A detailed tracker published by Medha Cloud found that across 166 major layoff events in the first months of 2026, middle management was among the clearest targets — with Amazon's restructuring alone shifting its manager-to-staff ratio from one manager for every six employees to one for every ten or more, explicitly framed as an anti-bureaucracy drive enabled by AI. These are not marginal adjustments. They are structural decisions with profound human consequences, being made at speed, and largely without the transition frameworks that the people affected deserve.

What makes this crisis particularly difficult to address is that many middle managers do not yet see it coming. They are being asked to champion AI adoption within their teams, to drive efficiency improvements, and to demonstrate value in an environment where the tools they are championing are simultaneously making their own roles harder to justify. There is a painful irony in being asked to implement the technology that is quietly replacing you.

IMPLICATIONS

For business leaders and boards, the message is direct: AI restructuring that touches middle management cannot be treated as a cost optimisation exercise dressed up in digital transformation language. These decisions require honesty, proper planning, and genuine investment in transition — whether that means retraining, redeployment, or respectful and transparent redundancy processes. Organisations that handle this badly will pay for it in trust, culture, and talent for years to come.

The human cost of this shift carries particular weight in the South African context. A report by SAP Africa found that every organisation surveyed expects an AI-related skills gap, with 38 per cent identifying the reskilling of employees as their top challenge and nearly half saying the same about upskilling. Critically, the report warns that shrinking training budgets at precisely the moment when reskilling demand is accelerating could lead to failed digital transformation initiatives and eroding competitiveness.

For South African organisations restructuring middle management around AI tools, this is a compounding risk: the layer of people being most directly affected is also the layer most responsible for driving the very capability-building that the transition requires. Removing that layer without a credible reskilling strategy in place is not just a human resources failure — it is a strategic risk with long-term consequences for organisational performance.

For middle managers themselves, the most important response is not resistance but reinvention. The roles that will survive and grow are those that lean into what AI cannot do: building trust across teams, navigating complex human dynamics, exercising ethical judgement, and providing the kind of contextual leadership that no algorithm can replicate. The managers who will thrive are those who become fluent in AI tools while doubling down on their distinctly human capabilities.

CLOSING TAKEAWAY

The crisis facing middle management in the age of AI is real, it is already underway, and it deserves an honest conversation at the highest levels of every organisation. AI is not a threat to be denied or a trend to be celebrated uncritically — it is a structural force that is quietly reshaping the architecture of how organisations work and who they need to work in them. For the managers caught in the middle, and for the leaders responsible for their futures, the time to plan is not when the restructuring announcement arrives. It is now, while there is still room to act with integrity, humanity, and genuine care for the people who have given their working lives to building the organisations that AI is now transforming.

Author Bio: Johan Steyn is a prominent AI thought leader, speaker, and author with a deep understanding of artificial intelligence’s impact on business and society. He is passionate about ethical AI development and its role in shaping a better future. Find out more about Johan’s work at https://www.aiforbusiness.net