Robots & Robotics

Disney and AWS Build Star Wars Robots That React to Guests

Disney’s Anzellan prototypes use AWS AI to choose reactions, prerecorded dialogue and animations in real time, with a public showcase planned at re:Invent.

By Rachel Whitman | Edited by Daniel Krauss Published: Updated:
Disney and AWS Build Star Wars Robots That React to Guests
Disney and AWS are developing interactive Anzellan robots that select dialogue and reactions in real time for character encounters. Photo: Disney

Key Notes

  • Disney Imagineering has introduced three Anzellan robotic-character prototypes that respond to visitors and each other using AWS-powered decision-making.
  • The system selects prerecorded, Imagineer-written dialogue and coordinates animations in real time, rather than generating unrestricted conversations.
  • The characters are scheduled for AWS re:Invent in Las Vegas from November 30 to December 4, while a regular theme-park rollout has not been announced.

Disney and Amazon Web Services have introduced a new kind of Star Wars character encounter: three small robotic Anzellans that react to visitors and one another in real time. The prototypes combine physical animation with AI decision-making, letting an interaction change according to what happens around them.

Walt Disney Imagineering R&D introduced Vilo Vele, Kado Bre’a and Jibu Bre’a on November 18. Disney says the Audio-Animatronics characters choose responses without a puppeteer operating them, extending its work on responsive robotic personalities.

AI Selects the Performance Within a Written Character

AWS technology powers the characters’ real-time intelligence. Disney describes fast-decision models for quick reactions and larger models for broader context. The figures can respond to gestures and speech, while coordinating their behavior with the other characters in the scene.

The important creative boundary is the dialogue. The system selects prerecorded audio and pairs it with dynamic animation; Imagineers write the lines and design the personalities. AWS’s announcement describes that same arrangement. The result is adaptive timing and response selection within a crafted performance.

That architecture gives the team a way to preserve recognizable characterization while allowing encounters to vary. A visitor’s wave can change the next response without requiring a model to invent a new line of dialogue. The physical performance and the choice of when to use it become separate parts of the control problem.

The Robots Have to Respond as a Group

A three-character scene creates demands beyond a single robot answering a question. Looking toward a visitor, waiting while another character speaks and reacting to that character all affect whether an encounter feels coherent. Animation, sound and decision timing need to work together.

Disney says it has optimized the components for low latency, but it has not released response-time measurements or detailed model specifications. The public announcement therefore describes the intended behavior and architecture, rather than a benchmark that can be compared directly with another social robot.

A Conference Showcase Comes Before a Park Rollout

The Anzellans are scheduled to appear at the Customer Showcase at Caesars Forum during AWS re:Invent in Las Vegas, running November 30 through December 4. Disney has not announced when or where the prototypes might become a regular theme-park experience.

The project complements the operational robotics explored in our report on Disney’s robotics collaborations. Those projects investigate parking support and backstage handling; the Anzellans put the interaction itself at the center of the robot’s job.

Our coverage of Hasbro’s character partnership follows another effort to combine licensed personalities with physical robots. Disney’s latest prototypes offer a concrete example of how authored storytelling can be connected to autonomous response selection, with the upcoming showcase providing the next opportunity to see that combination in use.

Disclaimer: RobotsBeat is an independent media brand owned and operated by NuvexMedia LLC, publishing news, research, and insights on artificial intelligence, emerging technologies, automation, and related industries. NuvexMedia LLC invests in and collaborates with companies across the AI, Robotics, technology, software, and digital innovation sectors. These relationships do not influence RobotsBeat's editorial coverage, and the publication maintains full editorial independence to provide accurate, timely, and objective information. © 2026 NuvexMedia LLC. All rights reserved. This content is for informational purposes only and should not be considered legal, tax, investment, financial, or other professional advice.

Artificial Intelligence (AI), News, Robots & Robotics