Values Statement: We believe AI should cause no harm, but enhance the quality of human life, by proactively adopting our AI Governance framework.

Evidence-based Transparent For governance

AI Incidents

Lawsuit Accuses Oracle's Algorithmic Data Processing System of Unlawful Data Collection and Privacy Violations - Highlighting the Importance of Responsible AI Governance

Dec 22, 2014

A lawsuit has been filed against Oracle, claiming that their automated system using algorithmic data processing was unlawfully collecting pe...

Facebook's Year-in-Review Algorithm: Showcasing Unwanted Memories - Harm Prevention in AI Incident

Dec 9, 2014

An incident involving Facebook’s ‘Year in Review’ algorithm highlighted painful and unwanted memories to users, including the death of a fam...

AI System Allegedly Terminates Employee Without Notice: A Case Study in AI Governance

Oct 18, 2014

Exploring an unsettling incident where an employee was terminated by an AI-powered personnel system, raising important questions about trust...

Uber's Secret Program 'Greyball': A Case Study on AI Governance for Trustworthy and Safe Ride-Sharing

Oct 1, 2014

Explore the instance of Uber's clandestine program 'Greyball', designed to bypass ride requests from known regulators. This AI incident high...

Alleged Data Quality Issues and Discrimination in USCIS's ATLAS AI Software: A Threat to Trustworthy Immigration Vetting

Aug 26, 2014

The controversial ATLAS software used by the US Citizenship and Immigration Services (USCIS) has come under fire for its opaque algorithmic...

Uber's Autonomous Vehicles Violating Traffic Laws: Running Red Lights in SF

Aug 15, 2014

An AI incident involving Uber self-driving cars running red lights during street testing in San Francisco highlights the importance of safe...

Starbucks Employees Allege Financial Instability Due to Kronos Scheduling Algorithm: A Call for Responsible AI Governance

Aug 14, 2014

Kronos's scheduling algorithm, when used by Starbucks managers, allegedly caused financial and scheduling instability for wage workers. This...

German Volkswagen Plant AI Incident: Robot Causes Fatal Accident Highlights Need for Responsible AI

Jul 15, 2014

A tragic incident occurred at a German Volkswagen plant when a robot, in the absence of safe and secure AI guardrails, pinned a worker to a...

Allegations of Influencing Gig Worker Pay through AI Algorithms: The Need for Responsible AI Governance

May 8, 2014

A multiyear ethnographic study suggests Amazon and Uber used algorithmic systems to manipulate wages for gig workers, offering lower pay for...

Inaccurate ALPR Reading Leads to Unwarranted Traffic Stop: Understanding AI Incidents and the Need for Responsible Governance

Apr 16, 2014

An instance of misreading by an Automated License Plate Reader (ALPR) camera incorrectly flagged a vehicle as stolen, leading to an unwarran...

Facebook's Automated Advertising Contains Anti-Semitic Categories: A Case for Responsible AI and Harm Prevention

Mar 4, 2014

An incident involving Facebook's automated advertising system highlighted the need for trustworthy AI. The categories generated using users'...

Google Nest Smoke Alarm Incident: Unintended Silencing of Legitimate Alarms

Jan 21, 2014

Recent testing on Google Nest Protect: Smoke + CO Alarm revealed an issue with the Nest Wave feature, which may inadvertently silence authen...

Data source

Incident data is from the AI Incident Database (AIID).

When citing the database as a whole, please use:

McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.

Pre-print on arXiv · Database snapshots & citation guide

We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.