Is Your Boss Actually an Algorithm?

A software developer starts her day by logging into the company’s project management system. An AI-powered notification immediately populates her task list, prioritized by an algorithm that has analysed project deadlines, and her past performance patterns. Throughout the day, the system tracks her commits, code quality metrics and time spent on each ticket. It automatically flags her when she falls behind the predicted pace for a task. 

Her calendar is optimized by another algorithm that schedules her meetings and even her relaxation time. Performance reviews are generated from aggregated metrics tracking her output against benchmarks. When she has questions about priorities, she interacts with a support chatbot. Weeks can pass without any meaningful interaction with her human manager.

Sound dystopian? This scenario isn't science fiction. Parts of it are playing out in different industries already, and if you think what happens in a warehouse or a coffee chain could never happen to a developer, you’d be wrong.

Welcome to the age of algorithmic management.

It’s the use of digital algorithms (increasingly, AI) to perform what managers traditionally do: assign tasks, monitor performance, provide feedback, resolve disputes and make decisions. The rise of algorithmic management represents one of the most profound shifts in how organizations function since the emergence of scientific management a century ago.

But how do you know if algorithms, rather than humans, are truly calling the shots in your workplace? Your boss might still be human, but their behaviour may not.

5 Signs Your Boss Is an Algorithm

The transition from human to algorithmic management often happens gradually. Here are some tell-tale signs:

1. Automated task assignment: Your work arrives through systems that optimize for efficiency rather than considering your preferences or development needs.
2. Continuous digital monitoring: Your activities are tracked in granular detail — commits counted, response times measured, or screens monitored — with limited transparency about what metrics matter or why.
3. Standardized feedback loops: Evaluations come as automated reports comparing your “numbers” against benchmarks, with little room for context or explanation.
4. Reduced managerial discretion: When you request exceptions or accommodations, you’re told “the system doesn't allow for that” or “we need to follow the process.”
5. Opaque decision rationales: You notice patterns in decisions affecting your work, but the reasoning behind them is rarely explained, creating a sense that invisible rules govern your work life.

Working for a traditional bureaucracy can also feel similar, but the digitization is what puts that feeling on steroids. These signs point to a fundamental shift that brings both opportunities and risks.

To understand why algorithmic management often feels dehumanizing, we need to look deeper at three distinct problems it creates.

More on AIDeepfake Phishing: Is That Actually Your Boss Calling?

3 Problems With Using Algorithms for Management 

1. Inaccuracy — When AI Lacks Human Context

Given the same data, it's increasingly unlikely that a human can outperform an algorithm in extracting useful patterns from it. But a lot of data relevant to a situation that humans have (we call it “context”) isn’t accessible to algorithms. This is not merely a technical limitation but a fundamental challenge of translating human experience into data.

Consider performance evaluation. An algorithm might track a product manager’s output in terms of features shipped, documentation produced or meetings attended. But it misses crucial context: the difficult stakeholder who required extra diplomacy, the junior team member who needed mentoring, or the technical debt that was skilfully managed behind the scenes.

This context blindness isn't necessarily about bias, it’s about the absence of a type of data that humans process naturally but algorithms cannot yet fully capture.

2. De-skilling — The Erosion of Human Judgment

When algorithms take over decision-making, humans gradually lose opportunities to develop and maintain their own judgment capabilities. This de-skilling effect has profound long-term consequences.

Software developers who rely heavily on AI coding assistants and shift solely to code evaluation may find their ability to solve complex problems independently diminishing over time. Financial analysts who depend on algorithmic recommendations might lose the intuition that once helped them spot market anomalies. Customer service representatives following AI-scripted responses risk losing the empathy and improvisational skills that define excellent service.

This phenomenon can create a dangerous cycle: as humans defer more decisions to algorithms, their skills atrophy, making them increasingly dependent on algorithmic guidance. The skill ladder (the progression from novice to expert through hands-on experience) begins to miss crucial rungs.

3. Undermining Organizational Context

Perhaps most insidiously, algorithmic management often conflicts with fundamental human needs that make work meaningful and motivating. Researchers have shown that these include autonomy, relatedness, competence, fairness, purpose and novelty.

Algorithmic management frequently undermines these dimensions:

1. Autonomy: When algorithms dictate exactly how and when you work, your sense of agency and control diminishes. Engineers may feel like "code-producing machines" when their work is scheduled and evaluated entirely by algorithms.
2. Relatedness: The human connection between team members weakens when mediated by algorithms. If my AI drafts emails for your AI to read, summarize and respond to, are we connecting at all?
3. Competence: Skill development suffers when algorithms optimize for narrow metrics rather than holistic growth. A sales team managed purely by conversion metrics might miss opportunities to develop consultative selling abilities.

These can translate into tangible business problems: reduced trust, lower intrinsic motivation, strategic compliance (gaming the metrics), increased turnover, and as word gets around, inability to hire well. Companies with high reliance on algorithmic management may find themselves paying premium salaries to compensate for diminished employee satisfaction, essentially buying compliance rather than earning commitment, and hoping that is enough. It rarely is. 

More on AIWhat Is Long-Thinking AI and Why Does That Matter?

Finding the Human-Algorithm Balance

The solution isn't rejecting algorithms but using them in a manner that keeps organizations human-centric. As I explain in “Re-Humanize”, that means investing in human skills and building organizational contexts that meet employee’s preferences for autonomy, competence and relatedness in the workplace (remember: people value these differently).

Some of the ideas on how to do this include: 

1. Use algorithms as smart coordination rules that facilitate collaboration rather than as surveillance systems.
2. Apply ensembling approaches where humans and algorithms work on the same tasks in parallel rather than divide up tasks into what the human and machine can each do better. 
3. Enable algorithm-assisted self-selection where systems help workers choose tasks matching their skills while also ensuring all necessary work gets done.
4. Practice decentralized designing by allowing teams to participate in shaping how algorithms integrate into their workflow, rather than imposing standardized solutions from above. 

Algorithmic management can improve on traditional management by reducing biases, and improving coordination, for instance, and it would be foolish to ignore these benefits. But organizational performance and strategic differentiation, also comes from the skills and motivation of humans. It would be unwise to forget that. The challenge isn’t to resist AI but to design workplaces where algorithms enhance human work rather than erasing its humanity. After all, organizations exist to serve us, not the other way around.