Carnegie Mellon Tests Human-Robot Interaction with Trash-Talking Gamebot

A trash-talking, Scrabble-playing gamebot with a goofy back story and an animated face is helping researchers at Carnegie Mellon University understand more about how people and robots interact. Victor the robot was developed in the university's Robotics Institute as one of a series of "social robots" to explore human-robot engagement.

The robot consists of a plastic torso topped with an old fashioned monitor-shaped head in which a video screen displays a talking head. He's seated between two computer displays on one side of a table touchscreen showing a virtual Scrabble board. Up to three players can take seats around the table to start the game. Because Victor has no arms, his word tiles are laid out digitally; the other players use touch to move their virtual tiles out onto the board. Victor has a voice, but those who want to converse with him must use a keyboard.

The setup resides near a coffee bar in a student center. Victor is available for play each day for a couple of hours with a research team member hovering nearby to handle problems as they surface. The goal is to allow him eventually to be available to play at any time without supervision.

Victor was created by a team that included collaborators from robotics, computer science, drama, design and entertainment technology. The work was led by Research Professor Reid Simmons.

The robot's backstory, written by Michael Chemers, now an associate professor of theater arts at the University of California, Santa Cruz, explains that Victor is "the precocious son of a pair of Detroit industrial robots"; he's attending the university on a Scrabble scholarship. Chemers also wrote much of his dialogue and maintains Victor's Facebook page, which offers wry commentary on all aspects of robotic life, such as this: "Do you ever have the desire to have the people who built you take you completely apart, dunk each piece in nonoxynol-9, and then reassemble you exactly the same way? Come on, we all feel this way from time to time, do we not? Anyone? Anyone?"

Victor's personality is, said Simmons, "edgy enough to be engaging, but not so much that people don't want to play." When Victor is losing a game of Scrabble, for example, he becomes angry and his mood changes, his expression changes, and what he says changes. "He's much more sarcastic," Simmons noted. "What surprised me is how strongly people react to that. When he starts trash-talking them, they start trash talking back."

And lose he does. The researchers have set his vocabulary at about 8,600 words, compared to the Official Scrabble Player Dictionary's, which features 178,000 words.

When he's playing, the robot will bow his head or look another player in the eyes. A light over his heart will change color and throb at different speeds depending on his mood.

Although the other players can't see Victor's tiles, they can see each other's. What the researchers have found is that the humans will "gang up" with each other when they play against the robot.

Explains Simmons, "They don't care if they lose to each other, but they don't want to lose to a machine."

About the Author

Dian Schaffhauser is a former senior contributing editor for 1105 Media's education publications THE Journal, Campus Technology and Spaces4Learning.

Featured

  • glowing data flow into abstract large language model

    White House Intros Classified Cybersecurity Review for Frontier AI Models

    The White House is moving forward with a voluntary framework for testing the cybersecurity capabilities and risks of advanced artificial intelligence models, though the standards guiding those evaluations remain classified.

  • cloud computing circuit board background

    Report: Cloud Crime on the Rise as Attackers Exploit Trust

    CrowdStrike's 2026 Threat Hunting Report traces a multiyear shift from conventional intrusions toward attacks that exploit trusted identities, cloud services, AI systems and software dependencies.

  • Blurred silhouettes of business people in a modern office with a glowing blue network overlay

    Open Secure AI Alliance Moves to Linux Foundation

    The Open Secure AI Alliance has moved under the Linux Foundation, giving the initiative what the organizations describe as a neutral home for developing open source tools, shared standards and defensive practices. The Alliance was launched by NVIDIA in July to develop open security technologies for AI systems and agents.

  • Abstract neural network 3D illustration

    Intel® AI EmpowerED: The AI-Ready Campus, Delivered

    Artificial intelligence is transforming higher education, prompting institutions to rethink how they manage infrastructure, security, governance, and workforce readiness. Successful adoption requires a strategic, institution-wide approach that aligns AI initiatives with educational goals, faculty enablement, and scalable operational frameworks.