According to the International Federation of Robotics (IFR), there were around 4.7m industrial robots operational worldwide as of 2024—just 177 for every 10,000 manufacturing workers. Annual installations surged during the pandemic-era automation frenzy but flattened off afterwards, with 542,000 installed in 2024. The IFR forecasts 619,000 installations in 2026.
Robotic arms that once moved along three axes typically now move along six, guided by sensors and cameras. A single robot is often able to perform several manufacturing steps. They have plunged in price as production has scaled up and Chinese suppliers have entered the business.
Robots were once rigidly designed for one activity. Now the machines can be reprogrammed for another job with a tweak to their code. This flexibility has improved the lifetime return on investment. General Motors became the first company to install an industrial robotic arm in the mid-1960s.
Advances in generative AI promise to close the "sim-to-real gap"—the problem that precise modelling of a robot's actions was often impossible owing to the many variables involved, with simulations tending to break the moment lighting or the shape of an object changed. Supersized AI models, trained on vast amounts of sensor and camera data, may allow robots to perceive, understand and react to physical situations much as a human would.
Jensen Huang of Nvidia declared on January 5th 2026 that "the ChatGPT moment for robotics is here." At the Consumer Electronics Show that week, Nvidia unveiled a suite of chips and freely available AI models designed specifically for robots. In October 2025 SoftBank announced it would acquire the robotics division of ABB, a Swiss industrial giant.
The future isn't what it used to be. (It never was.)