Managers were disproportionately affected by the waves of post‑COVID tech layoffs that began in late 2022. Large companies such as Meta, Google, and Amazon popularised phrases like “flattening the org” and “reducing bureaucracy,” which in practice meant cutting management layers that expanded during the 2021–2022 hiring sprees. In hindsight, that move can appear reasonable: AI models can already automate scheduling, draft performance reviews, coordinate cross‑team communication, and assist with prioritisation and decision support—tasks historically performed by managers. In experimental extremes this logic has produced reporting ratios like 50 individual contributors to one supervisor.
The simple conclusion: because AI can or will soon handle many managerial tasks, fewer managers are needed; decision‑making can be distributed as individual contributors become better at orchestrating agentic workflows and exercising judgment without close managerial oversight. In this view, everyone effectively becomes a manager.
Why that narrative is flawed
Organisations are cutting managers just as those roles are becoming critical to realising AI investments. Several recent data sources support this. Microsoft’s 2026 Work Trend Index Annual Report concludes that “organisational factors—culture, manager support, talent practices—account for twice the reported AI impact of individual effort alone.” After leadership sets AI strategy and incentives, “it’s managers who operationalize it, and the data shows the impact of their ability to do so.”
Microsoft’s data show concrete effects: when managers actively modelled AI use, employees reported a 17‑point lift in perceived AI value, a 22‑point increase in critical thinking about their AI use, and a 30‑point increase in trust in agentic AI. When managers created psychological safety for experimentation, employees reported up to 20 points higher AI readiness and value—and were 1.4 times more likely to be high‑frequency users of agentic AI.
The manager effect is even stronger among more advanced AI users — Microsoft’s “Frontier Professionals,” who make up 16% of respondents and use agents for multistep workflows and build multi‑agent systems. This group was more likely to report that their manager uses AI (85% vs. 64%), sets quality standards for AI work (83% vs. 57%), encourages experimentation (84% vs. 61%), and rewards work redesign regardless of outcome (26% vs. 11%). Microsoft notes that “in many cases, employees are moving faster than the organization around them,” and managers are the layer that helps resolve this “Transformation Paradox” by translating organisational strategy into team practices that let individual AI work produce value.
The role is expanding, not contracting
One could argue that once AI adoption is widespread and managers no longer need to manage the change, many aspects of the role will be automatable and the role will contract. But if contraction were already occurring, early signs would be visible — and the data show the opposite.
The LeadDev Engineering Leadership Report 2026 surveyed 600 engineering leaders, 55% of whom are engineering managers or managers of managers. The report states that “AI is simultaneously expanding what leaders can do technically and what is expected of them organizationally, without reducing the demands on their time in either dimension.” Managers are becoming more technically hands‑on and report growing responsibilities:
- 63% say their scope and area of responsibility increased over the past 12 months.
- 60% saw increased communication with team members, customers, and stakeholders.
- 22% have more teams reporting to them.
- 29% have more direct reports.
Respondents most often cited increased time spent on architectural decisions and technical strategy.
One interpretation is that more teams and direct reports reflect intentional flattening working as intended from a business standpoint. Another — not mutually exclusive — interpretation is that the role is in transition: managers are doing old work at greater scale while also taking on new work to make AI a core team practice. Either way, this is not evidence of contraction. Contraction would mean the role’s scope is shrinking as AI and individual contributors take on more of the work; instead, flattening produces more teams and more reports, not manager elimination.
What decision‑makers should consider
This does not mean organisations should stop scrutinising reporting lines or removing genuinely harmful layers of bureaucracy that slow decision‑making. But it does mean asking a tougher question before the next wave of cuts: are you reducing management based on what managers used to do, or based on the critical work they are doing now and will need to do next?
Treating managers as overhead on the basis of past duties discounts the emerging evidence that managers today act as infrastructure: the critical layer that translates AI investment into actual, team‑level value. Flattening on the assumption that AI will self‑facilitate adoption or that unguided individual effort will generate value is a productivity bet the data do not support.
