AI isn't just automating tasks anymore: it's eliminating entire management layers, giving leaders the capacity to oversee significantly larger teams and compressing corporate hierarchies in the process.
Corporate America has spent the last decade thickening its org charts, inserting layer upon layer of middle management between the CEO and the front lines. McKinsey senior partner Alexis Krivkovich puts it plainly: companies added between one and three management levels over the past ten years, bogging down decision-making and bloating payrolls. Now, the firm sees artificial intelligence as the tool to reverse that trend, and they're not alone. IBM, Factory, and a growing cohort of enterprise technology companies are actively restructuring around the idea that AI agents can absorb enough operational overhead to flatten teams and accelerate how fast work gets done.
Krivkovich told The McKinsey Podcast that AI gives leaders "more of a superhuman capacity to manage across bigger scopes," which directly enables companies to remove organizational bloat and move faster. The core argument is straightforward: managers currently spend an outsized portion of their time on coordination, status updates, approvals, and resource allocation. If AI agents handle those functions, one leader can do the work that previously required three or four middle managers acting as information conduits.
What does this look like in practice? IBM's consulting arm, which employs roughly 150,000 human consultants, is already building what senior vice president Mohamed Ali calls "digital workers" into its project teams. Ali, who spoke with Business Insider about the shift, expects entirely new management structures to emerge. His key insight is that these digital workers won't be managed like people. Instead, systems will set guardrails, allocate tasks, and monitor output autonomously. That's a fundamentally different operational model than asking a human manager to approve timesheets and route assignments.
At Factory, an AI-native software development platform that counts Nvidia, Adobe, and EY among its clients, the vision is even more compressed. Cofounder and CTO Eno Reyes predicts that org charts will condense horizontally, with AI agents picking up the coordination and execution load that previously required dedicated personnel. The platform already deploys autonomous coding agents for consulting engagements, giving a concrete example of how AI collapses the distance between a strategic brief and a finished deliverable.
The consulting industry, in particular, stands at a turning point. Firms have historically operated on a leverage model: partners oversee directors, who oversee managers, who oversee associates, who oversee analysts. Each layer adds margin and cost. If AI agents can perform the analysis, generate the presentations, and even draft the client communications, the economic rationale for maintaining deep pyramids weakens considerably. Some observers have started calling this shift "The Great Flattening," and the label fits. We are watching the restructuring of corporate hierarchies in real time.
Who Gets Squeezed and What Comes Next
The implications vary by sector. In life sciences, McKinsey envisions "squads of agents" accelerating research and development cycles by handling data synthesis, literature reviews, and preliminary analysis. In back-office functions like human resources, finance, and legal, AI agents can already automate routine workflows and reallocate resources dynamically, reducing the need for dedicated coordinators and team leads.
But flattening carries real risks. Remove too many layers too quickly and you lose institutional knowledge, mentorship pipelines, and the human judgment that catches errors algorithms miss. A 2023 study from Harvard Business School found that middle managers spend roughly 40% of their time on coordination tasks that could theoretically be automated, but they also serve as critical buffers between executive strategy and frontline execution. The companies that navigate this transition well will likely be those that retrain displaced managers into higher-value strategic and client-facing roles rather than simply cutting headcount.
The economic incentives, however, are hard to ignore. Research from Gartner projects that by 2026, organizations that have redistributed management responsibilities to AI will see a 25% reduction in operational costs. For startups and scaling businesses, the lesson is even sharper: you may not need to build those layers in the first place. AI-native organizational design, where autonomous agents are embedded from day one, could let smaller teams punch well above their weight without ever developing the corporate bloat that legacy firms are now trying to shed.
