A grocery store robot developed by Heriot-Watt University was trialed at a Scottish supermarket, where it failed to provide helpful assistance to customers and exhibited behavior that caused shoppers to avoid specific areas of the store. The robot's performance in handing out samples was significantly lower than that of human employees, as customers were reportedly suspicious or uncomfortable interacting with it.
Heriot-Watt Univeristy in Scotland developed an artificially intelligent grocery store robot, Fabio, who provided unhelpful answers to customer's questions and "scared away" multiple customers, according to the grocery store Margiotta.
Risk classification
- Primary risk domain: 7 AI system safety, failures, & limitations
- Primary risk subdomain: 7.3 Lack of capability or robustness
The robot failed to perform its tasks effectively, providing unhelpful, overly general directions and failing to successfully interact with customers to distribute samples.
Causal factors
- Entity: AI
- Intent: Unintentional
- Timing: Post-deployment
The incident was caused by the AI robot's unhelpful interactions and behavior during its post-deployment trial at the supermarket, which was an unintentional outcome of its programming.
EU AI Act risk tier
- Risk tier: 4 Minimal or No Risk
Risk Level 4: Minimal or No Risk. The grocery store robot is used for simple customer service and entertainment applications, posing low or negligible risks to users and society.
AI system and alleged parties
- AI system: Fabio (Heriot-Watt University)
- AI purpose: Chatbot; Robotic Manipulation
- Behaviour type: Autonomous
- Alleged developer: Heriot-Watt University
- Alleged deployer: Margiotta, Heriot-Watt University
- Alleged harmed parties: Store Patrons
Harm severity
Highest direct severity in any category: Negligible. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.
- Physical: direct Negligible, indirect Negligible
- Infrastructure: direct Negligible, indirect Negligible
- Property: direct Negligible, indirect Negligible
- Financial: direct Negligible, indirect Negligible
- Environmental: direct Negligible, indirect Negligible
- Malicious content: direct Negligible, indirect Negligible
- Differential treatment: direct Negligible, indirect Negligible
- Civil rights: direct Negligible, indirect Negligible
- Democracy: direct Negligible, indirect Negligible
- Privacy: direct Negligible, indirect Negligible
- Psychological: direct Negligible, indirect Negligible
- Epistemic: direct Negligible, indirect Negligible
- Child sexual exploitation and abuse: direct Negligible, indirect Negligible
People affected
- Occurrences reported: 1
- People reportedly harmed: 1
- People reportedly exposed: 50
Potential causes
Management
- Premature Commercial Trial: Supermarket trialed the robot before its practical utility was verified.
Technology
- Vague Dialogue Responses: The remote NLP system gave overly generic answers like 'beer is in alcohol'.
- Unappealing Physical Presence: The robot's physical form or behavior scared customers away from sausages.
Data Inputs
- Lack of Store Layout Data: The AI lacked specific aisle mapping data to provide helpful directions.
Human Factors
- Customer Suspicion of Robots: Shoppers were uncomfortable interacting with a robot offering food samples.
Process and Methods
- Inadequate Real-World Testing: The conversational AI was not tested in a noisy, busy retail environment.
- Poor Task Reassignment: Deploying a conversational robot to hand out food samples was a poor fit.
Information quality
- Classification confidence: High
- Reason for confidence: The report clearly describes the robot's deployment, its specific failures (unhelpful directions, scaring customers), and the outcome of the trial. There is no ambiguity about the AI's role or the lack of severe harm.
An experimental grocery store robot trial in Scotland failed due to unhelpful customer interactions and poor performance. The incident was entirely commercial and localized, with negligible national security implications.
- Overall national security impact: Negligible
- Response level: Minor
- Scope: Single nation
- Primary target: United Kingdom
- Alleged perpetrator: Unknown
Threat characteristics
- Imminence: Long-term. The incident was a localized, brief trial in 2018 that was ended early, posing no ongoing threat.
- Autonomy: Full autonomy. The robot was deployed to interact with customers and hand out samples independently without direct human intervention.
- Novelty: Established threat. Poorly performing customer service robots and basic conversational failures are established technological limitations.
Impact by dimension
- Physical security: Negligible. No physical harm, kinetic threat, or damage to critical infrastructure occurred during this local supermarket trial.
- Information security: Negligible. The incident did not involve any information warfare, intelligence compromise, or systematic information manipulation.
- Sovereignty: Negligible. The incident was a localized commercial trial in a grocery store and had no impact on government functions or state sovereignty.
- Economic security: Negligible. While the supermarket experienced minor operational inefficiencies, there was no threat to strategic industries, financial systems, or national economic security.
- Societal stability: Negligible. No societal instability, human rights violations, or mass surveillance occurred; customers merely avoided a poorly performing retail robot.
.avif){kind=small}