Working for the Algorithm: The Quiet Ascent of the Machine Boss
Working for the Algorithm: The Quiet Ascent of the Machine Boss
This article explores how algorithms are quietly taking over management and decision-making in the workplace.
At 8:11 pm on August 29, 1997, Skynet became self-aware.
In the iconic Terminator films, this artificial intelligence system, created by Cyberdyne Systems for the U.S. military, quickly concluded that all human life posed a threat to its existence. Skynet's response was to launch a devastating nuclear attack, paving the way for its army of autonomous humanoid hunter-killer units to wage war against humanity.
This chilling vision of a machine uprising has captured the public imagination for decades. The idea that AI could one day turn on its creators and seek to dominate the world has spawned countless dystopian narratives, from Terminator to The Matrix. But the reality unfolding around us is far more subtle, and perhaps even more interesting.
Rather than a violent coup, the machines have already infiltrated our workplaces through a more gradual process – one where they've been invited in, first as tools to augment human workers, and then to actively manage and oversee them. The "machine-boss" has become the unseen overseer, quietly dictating the terms of our employment and the trajectory of our careers. Many of us are already working for this algorithmic overlord, whether we realize it or not.
As we'll explore in this article, the machine-boss's influence extends far beyond the warehouse floors and delivery routes. Its reach now encompasses the knowledge economy and the creator landscape, as algorithms increasingly make decisions about who gets hired, who gets promoted, and who gets rewarded. The machine-boss is here, and its power is only growing.
The Algorithm in the Warehouse
In our imaginations, the Amazon warehouse is a bustling, high-tech wonderland – a choreographed dance of human workers guiding robotic assistants, working in harmony to fulfill the relentless demands of the e-commerce giant's supply chain. But the reality is a little different.
As we step inside the walls of Amazon's distribution centers, we find a different dance unfolding – one where the robots are the leaders, and the human workers follow their lead. Every day, workers check into their jobs, ready to tackle the warehouse maze. These "pickers" are handed computers that dictate their every move, guiding them through the aisles. Much like the rats in a laboratory experiment, the humans are run through the stacks, tasked with filling boxes with toys, games, and protein bars at a breakneck pace.
The experience can be grueling, as one Amazon picker shared on Reddit: "I'm speed walking my a* off, in and out of the bins within seconds, to no avail. I'm young and fit. I seem to be walking even a little faster than most people I see moving through the aisles. This really doesn't seem worth $12/hr so far*."
To motivate these workers, Amazon has gamified the experience, introducing "FC Games" – a series of arcade-style mini-games that allow pickers to earn virtual rewards. Just as researchers have long used food pellets and other treats to condition rodents to run through complex labyrinths, Amazon has engineered its own system of virtual carrots. The "MissionRacer" game, for example, allows pickers to move a car around a track as they sort products into appropriate boxes.
But the fun and games quickly give way to the harsh realities of algorithmic oversight. The system tracks 'time off task,' which the company abbreviates as TOT. If workers break from scanning packages for too long, the system automatically generates warnings and, eventually, the employee can be fired.
The machines' influence extends beyond the warehouse walls, as Amazon's "computer overlords" also dictate the workflows of its delivery drivers. Video cameras, telematics devices, and smartphone apps monitor the drivers' every move, with software determining how many packages they should be able to deliver in a 10-hour shift – a number that keeps creeping up.
Just as a lab rat learns to navigate the maze with precision, lest it forfeits its reward, the Amazon pickers and drivers must dance to the tune of the computer, or risk facing the consequences. The carrot has been dangled, but the stick is never far behind.
The Gig Workers' Algorithmic Dance
The promise of the gig economy was one of freedom and flexibility – the ability to work when you want, take on as much or as little work as you please, and be your own boss. For many, the idea of becoming a delivery driver or rideshare operator held the allure of extra income and the autonomy to chart their own course.
But gig workers, too, have found themselves under the watchful eye of the machine-boss. Just as the Amazon warehouse pickers are guided by the relentless pace of the computer overlord, the delivery drivers for apps like DoorDash and Uber are subject to the whims of the same unseen hand.
The challenge for the platforms is immense – how do you efficiently match millions of gig workers with the endless stream of delivery requests and ride hails? The answer lies in the power of algorithms, which can take into account a multitude of factors, from the driver's location and rating to the estimated time of arrival, to orchestrate the perfect pairing.
As a new gig worker, you might expect to open your app to be ushered into an open auction market where you can spot the best pickup and take it. But the reality is that the model decides what you can see. The algorithm determines which delivery or ride requests you are even made aware of, based on its calculations of what will be the most efficient match.
These drivers have learned to position themselves strategically, hovering near popular restaurants and hotspots, hoping to be the closest option when the algorithm goes searching. As one Quora user observed, the signs of this algorithmic influence are all around us: "If you are in an area and see a bunch of people sitting in their cars (especially a plethora of Prius) or a group of people standing around holding their phones and it's obvious they aren't friends, than you have stumbled upon a group of drivers that are all waiting for a [Doordash] request."
The Algorithmic Grip on Knowledge Work
But the machine-boss's influence doesn't stop at the warehouse floor or the gig worker's car. That same unseen hand has also crept into the realm of white-collar employment and the creator economy.
Just as the machine-boss proved adept at efficiently matching delivery drivers with customer demands, the same principles have been applied to the hiring process. According to a study by Harvard Business School, 99% of Fortune 500 companies now use Applicant Tracking Systems (ATS) powered by artificial intelligence to sift through the endless stream of resumes.
This outsourcing has consequences. In 2018, Amazon realized that the hiring software it had been developing for four years was systematically scoring qualified female candidates below their male counterparts. The reason was simple: the AI had been trained on the company's own hiring history, which was heavily skewed towards men in the male-dominated tech industry. The algorithm had internalized this bias, excluding talented women from even being considered.
This story is not unique to Amazon. Across the United States, Germany, and the United Kingdom, 63% of companies rely on these AI-driven hiring tools to identify the "best" candidates. But as the Amazon example illustrates, the definition of "best" is often shaped by the historical data used to train these models – a troubling reality that can perpetuate existing inequities.
The influence of algorithms extends beyond the traditional workplace and into the creator economy as well. As more and more individuals turn to platforms like YouTube, TikTok, and Instagram to build their personal brands and earn a living, they find themselves at the mercy of the very same machine-driven forces.
Creators produce content and provide it to the algorithm, hoping to be rewarded with follows, likes, and advertising dollars. But the algorithm remains finicky and particular, constantly optimizing for the engagement metrics that drive the platform's success. A creator's ability to thrive in this environment is often determined not solely by the quality of their work, but by their ability to game the system and appease the algorithm.
Just as the delivery drivers learned to position themselves strategically, waiting near popular restaurants in the hopes of being selected, creators have developed their own techniques to boost their visibility. From incorporating trending keywords to leveraging influencer networks, the dance with the algorithm has become a necessary part of the creative process.
The Rise of the Machine Boss
The modern workforce is undergoing a transformation, as the influence of algorithms and artificial intelligence permeates every corner of the job market. From the warehouse floors (~2 million workers) to the gig economy (over 23 million Americans earning money through online platforms) and the creator landscape (45 million professional creators), the machine boss is steadily asserting its dominance.
These are not isolated pockets of technological upheaval. Rather, they represent the vanguard of a broader trend. After all, human resource expenses are one of the largest costs for most companies, as the modern economy has become increasingly service-oriented. And the machine boss offers a tantalizing solution – the ability to inspect and evaluate worker productivity with a level of detail that no human manager could ever hope to match.
Take the example of the sales profession, which employs 5.7 million individuals in the United States alone. Traditionally, sales managers have had to rely on reports and periodic check-ins to gauge the performance of their teams. But now, software like Gong can listen in on every single sales call, analyze the content, and provide detailed coaching notes on everything from the time spent speaking to the key points that were missed.
Today, this feedback is still delivered to human sales managers, who pass it on to their teams. But this is merely an interim step – the logical progression is for the machine to provide that feedback directly, cutting out the human middleman altogether. Just as Target's "checkout game" measures and reports on the performance of cashiers and checkout clerks, this kind of granular, data-driven performance tracking will become the norm across occupations.
The implications are clear: wherever there is a job that involves repetitive, data-driven tasks, the machine boss can out-inspect and out-manage a human manager. The machine can evaluate worker productivity with a level of detail and precision that no human manager could ever achieve. And as artificial intelligence continues to advance, even the most complex, creative roles may fall under the purview of the algorithm.
Outmaneuvering the Machine Boss
Just as the Terminator movies depicted a future where humanity fought back against the tyranny of the Skynet AI, the real-world rise of the machine-boss has also inspired acts of resistance and ingenuity from workers.
In the gig economy, for example, drivers for platforms like DoorDash have refused to simply acquiesce to the algorithm's demands. In 2021, over 21,000 DoorDash delivery drivers banded together in a Facebook group called #DeclineNow, using collective action to strategically decline low-paying orders. The goal was to force the algorithm to offer higher compensation rather than simply accepting the scraps it doled out.
Other gig workers have turned to third-party apps like Gridwise, SherpaShare, and Mystro, which help them better understand the algorithm's inner workings and find ways to game the system to their advantage. And in Jakarta, GoJek drivers have even resorted to hacking their phones, using GPS-spoofing apps to trick the algorithm into thinking they're located in high-demand areas, even when they're waiting nearby.
The rise of the machine-boss is just the latest in the long history of management techniques that workers have had to navigate. Back when sales managers used to track outbound calls, savvy sales reps would smile and dial fax machines to hit their quotas. And if AI-powered software like Gong starts monitoring sales calls, you can bet that workers will find their own creative ways to circumvent the machine-boss's scrutiny, whether it's through talking AI sock puppets or other workarounds.
The machine-boss may be a formidable foe, but the human spirit remains undaunted. if history is any guide, the resistance will find a way to fight back – one clever hack, one coordinated protest, or one creative workaround at a time.