Boston Dynamics Is Teaching Its Robot to Behave, Not Just Move

Key points

• Boston Dynamics is hosting a free webinar at 11:00 a.m. ET on Wednesday, 22 July 2026 on human-robot interaction design.
• The session focuses on Atlas, Boston Dynamics' electric humanoid, which is among the few humanoids currently in commercial trials.
• Speakers argue that predictability, not human-like appearance, is the key to safe robot-human coexistence.
• Large language models, the AI technology behind chatbots like ChatGPT, are helping robots move away from rigid, pre-programmed scripts.
• One speaker spent 14 years as a character animator at Disney Studios before joining robotics.

Humanoid robots are leaving the lab. They're showing up on factory floors and in commercial spaces, working alongside real people, and that shift is raising real tension: Hyundai workers walked off the job over it just last week.

Boston Dynamics thinks the answer to that tension is good design, not a friendlier face.

The company is running a free webinar called "The Art Behind Human-Robot Interaction" on Wednesday, 22 July 2026, at 11:00 a.m. ET. Two speakers will walk through how they're making Atlas feel predictable to work near. As first reported by The Robot Report, the session draws from both engineering and an unlikely source: classic animation.

Leland Hepler, a product designer at Boston Dynamics, spent 14 years animating characters at Disney Studios on films including The Lion King and Chicken Little, and a decade art directing AI characters for video games. He now applies those instincts to Atlas, using motion and visual cues to signal what the robot's about to do before it does it.

Think of it like a good driver who checks their mirror and signals before changing lanes. A small movement or visual indicator that says: I see you, and here's what comes next.

Joining Hepler is Mario Bollini, the director of human-robot interaction at Boston Dynamics, who guided Atlas through its launch at CES and helped scale the company's four-legged Spot robot to thousands of deployed units worldwide.

What will workers actually learn from this?

Attendees will get concrete answers on four things. First, how robots communicate intent through "anticipatory cues," small signals that tell nearby humans what the machine is about to do. Second, how the electric Atlas is designed to work alongside people without safety caging, the physical barriers usually required to keep workers away from industrial machines. Third, how large language models are replacing rigid, pre-scripted behaviour with something more flexible. Fourth, why leaning into a robot's mechanical appearance, rather than mimicking a human, actually reduces the "uncanny valley" effect, the unsettling feeling people get when something looks almost but not quite human.

That last point matters more than it sounds. A robot that looks human but moves wrong is creepy. A robot that looks like a machine and moves honestly is just a tool.

The webinar is free to register for at the Boston Dynamics website. No technical background is needed. If your team might share a floor with a humanoid robot in the next few years, it's worth an hour of your morning.