For generations, some of the greatest sporting achievements have been measured in seconds. Usain Bolt ran 100 metres in 9.58 seconds, setting a world record that has stood since 2009 and establishing a benchmark for human speed.
At the other end of the distance spectrum, Kenya’s Eliud Kipchoge redefined what seemed possible in marathon running when he completed the Berlin Marathon in 2 hours, 1 minute and 9 seconds in 2022, then a world record. In April 2026, fellow Kenyan Sebastian Sawe pushed that boundary even further, becoming the first athlete to officially complete a marathon in under two hours with a time of 1 hour, 59 minutes and 30 seconds. These numbers represent decades of human progress: extraordinary talent combined with years of training, conditioning, technique and endurance.
Today, we have numbers from a machine. HONOR humanoid Robotics D1 has demonstrated a 100-metre time of 9.32 seconds, faster than Bolt’s 9.58-second human world record. Its reported performances also include 400 metres in 39.45 seconds, 1,500 metres in 2 minutes 30 seconds and a half marathon in 50 minutes 26 seconds, reaching a peak speed of 14.5 metres per second. The immediate temptation is to frame this as another human-versus-machine contest. But for HONOR’s AI experts, the more interesting story is how artificial intelligence teaches a machine to move at that speed, how a robot learns differently from an athlete, and what that tell us about the next generation of intelligent devices.
The answer begins with understanding that running is far more complicated than simply moving two legs quickly.
How AI teaches a machine to move
Human beings learn movement almost without thinking about it. A child learns to stand, walk and eventually run through years of physical experience. Our brains constantly process information from our eyes, muscles, joints and surroundings, helping us adjust our posture and balance without consciously calculating every movement. For a robot, none of this is instinctive. Movement has to be learned in a much more deliberate way.
This is where AI becomes important. Rather than simply programming a robot with a fixed instruction to move one leg and then the other, AI allows the machine to learn patterns of movement and improve how it responds. The robot can be trained to understand what happens when its weight shifts, how much force is required for the next step and how to correct itself if its balance changes. The faster it moves, the more important those decisions become, because even a small error can affect the entire motion.
In that sense, AI is doing far more than helping the robot run quickly. It is helping the robot coordinate its body and make decisions continuously. Running becomes a combination of perception, balance, prediction and action, all happening in fractions of a second. This is what makes advanced robotics such an interesting demonstration of AI. It turns intelligence into something we can physically see.
Robots learn differently from humans
Humans learn movement through experience, repetition and instinct. We fall, adjust and try again. Over time, the brain builds what we often call muscle memory. An elite runner then spends years refining those movements through training, learning how to make small improvements in pace, stride, breathing and energy use.
A robot does not have human instinct, emotion or physical memory. Instead, it depends on data, sensors, algorithms and repeated training. It learns by identifying patterns and understanding what produces a successful outcome. When something changes, the system has to interpret the new information and decide how to respond.
That difference is important because it shows us what AI is increasingly becoming capable of. It is moving beyond recognising information and generating responses and learning how to interpret what is happening around it and then decide what action should follow. In robotics, that decision may mean changing a stride or adjusting balance. In other technologies, the same principle could help devices become more responsive and more useful to the people using them.
What robotics tells us about the future of smartphones
At first glance, a humanoid robot running at high speed may seem far removed from the smartphone in your pocket. Yet the two technologies are becoming connected by a similar idea: devices are gradually moving from simply responding to commands towards understanding context and taking more intelligent action.
Today, most people still interact with smartphones in a very direct way. We open an application, type a request, search for information or move manually from one service to another. AI is already beginning to change that experience. Phones are becoming better at understanding what users are trying to achieve, anticipating needs and connecting different tasks in ways that require fewer instructions.
Robotics gives us a glimpse of where that evolution could go. A robot cannot function effectively if it waits for a human to instruct every individual movement. It needs to understand its environment, process information and respond independently within certain parameters. Smartphones are unlikely to behave in exactly the same way, but the underlying principle is relevant. The future smartphone may increasingly act less like a collection of applications and more like an intelligent assistant that understands context, connects information and helps complete tasks.
Robotics is not necessarily separate from the future of personal technology. It can act as a testing ground for the kinds of AI capabilities that could eventually influence how smartphones and other connected devices understand and interact with the world around them.
Why the future of AI is becoming physical
For most people, AI still lives on a screen, helping us generate text, edit images, translate languages or answer questions. Robotics changes that by bringing intelligence into the physical world, where machines must understand their surroundings, make decisions and turn those decisions into action.
This is what makes embodied AI so significant. The next phase of AI will not only be about what machines can know or create, but what they can do. Robots are one expression of that shift, but the same thinking could influence smartphones, wearables, vehicles and other connected devices, making technology more aware of context and more capable of responding to it.
A fast humanoid robot, therefore, is about more than beating a human record. It shows what becomes possible when AI can understand movement, make decisions in real time and act in the physical world, capabilities that could ultimately reshape the devices we use every day.