No organization in the world operates more robots than Amazon, according to a detailed Ars Technica report published October 12. The company has more than one million robots working in its warehouses and plans to grow the fleet by building new factories for more capable machines. It is also the second-largest corporate employer in the world, with more than 1.5 million full-time and part-time workers, based on its 2025 annual report.

Robots at scale

Amazon’s fulfillment network handles tens of millions of same-day and one-day package deliveries each day, and the robots move shelves, carry totes and handle other repetitive tasks around the clock. Amazon executives, Ars reports, say the growing fleet eases the burden on human staff by taking on the toughest tasks, those with the greatest risk of physical fatigue and injury. Warehousing has long been a hazardous sector: according to the US Occupational Safety and Health Administration, as cited in the report, the injury and illness rate for US warehouse workers has typically been more than double the average across all industries.

Robots remain far from able to do everything people do. Handling the huge variety of items in an order, packing mixed products and dealing with damaged goods still rely on human workers, so Amazon keeps a very large human workforce working in tandem with its machines. Its promise of fast delivery to around 200 million Prime subscribers requires hundreds of thousands of employees to assemble and ship orders quickly.

What researchers are worried about

Labor advocates and independent researchers cited by Ars say the picture is not all upside. Brad Greenwood, a business professor at George Mason University, argues that historic automation mostly happened away from workers, taking high-risk tasks out of human hands. “The rise of collaborative robots really changes that dynamic,” he said, for two reasons: robots are physically close to workers, and “the robot has the ability to dictate the pace of work.”

That second point is the core of the concern. When a machine feeds work to a person, the speed of the system can set the speed of the human, and a pace that suits a robot may be physically punishing for a body. Researchers also note that the sector is characterized by very high turnover and injury rates, with almost two million people working in US warehousing. Whether adding robots improves or worsens those numbers depends on how the work is designed, a question that independent studies, not company statements alone, need to answer.

Why it matters

Amazon’s fleet is the largest real-world test of how AI-driven automation reshapes a workplace. The company is also feeding the same trend that is visible in robotics research, where “foundation models” for robots are meant to let machines handle more varied tasks; our explainer on vision-language-action models covers that technical side, and our guide to humanoid robots covers the human-shaped machines some companies hope will fill the gaps. As robots take on more tasks, the human roles that remain tend to be the harder-to-automate ones, which may mean fewer jobs overall, different jobs, or both.

Policy implications are open. Regulators and unions are asking whether pace-setting algorithms should be limited, whether robot-related injuries should be tracked separately, and how to retrain workers whose tasks disappear. Ars does not report a specific new rule, and Amazon has not published a plan to cut its workforce as part of the story.

Broader signals

The same week brought a small, vivid sign of public friction around automation: The Verge reported that a woman in Philadelphia was arrested and charged with disorderly conduct after standing in front of an Uber delivery robot to protest robots taking human delivery jobs. That is a single incident, but it fits a wider pattern of debate over robotaxis, delivery robots and AI agents and their effects on lower-skill work. For background on the logistics sector, see our piece on how AI is used in logistics and supply chains.

What to watch next

  • Amazon’s disclosures on fleet size, new robot models and the factories planned to build them.
  • Independent injury and pace-of-work studies in highly automated facilities.
  • Any proposed labor rules on algorithmic work pacing in warehouses.
  • Whether Amazon’s headcount trends diverge from its robot count.
  • Public responses to delivery robots in cities where they are being rolled out.

Key takeaways

  • Amazon has more than 1 million warehouse robots and over 1.5 million human employees.
  • Executives say robots take over the most strenuous tasks; researchers say close-proximity robots can set a demanding pace.
  • US warehouse injury rates have typically been more than double the all-industry average.
  • Robots still cannot handle the full variety of warehouse work.
  • Evidence on net effects for workers is still being gathered.

Frequently asked questions

Are robots replacing Amazon’s warehouse workers?

Amazon still employs over 1.5 million people and relies on humans for many tasks. The long-term effect on headcount is an open question.

What is a collaborative robot?

A robot designed to work physically close to people, rather than behind safety cages, which changes both safety and the pace of work.

Is the robot count a measure of AI progress?

Not directly. Many warehouse robots follow fixed routines; the trend to watch is how much AI lets them handle more varied tasks.