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

Controversial AI Software Company Enacts Changes Towards Responsible AI

In response to negative publicity, a software company has initiated modifications aimed at promoting trustworthy AI and safe practices. This...

Analyzing the COMPAS Recidivism Algorithm: A Step Towards Trustworthy AI

This AI incident maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM). Understanding the workings of the COMPAS r...

Debiasing Word Embeddings: A Step Towards Responsible AI and Harm Prevention

Exploring the importance of debiasing word embeddings to promote fairness in AI. This AI incident maps to the Govern function in HISPI Proje...

Google's Anti-Internet Troll AI Platform: A Test of Safe and Secure AI - Highlighting the Importance of Trustworthy AI

Recent research reveals potential vulnerabilities in Google's anti-internet troll AI platform. This incident underscores the need for respon...

AI Opinions on Homosexuality: A Case Study for Safe and Responsible AI Governance

Google's AI case highlights the importance of trustworthy AI, underscoring the need for robust governance mechanisms to prevent harm. This A...

Amazon's Algorithm Adjustments: A Case Study on Promoting Trustworthy AI for All

This AI incident maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM). Amazon's decision to modify its search and...

Google Apologies for Racist Auto-Tagging Incident in Photo App: A Case Study in Responsible AI

This AI incident, involving Google's photo app mislabeling images of people of color with offensive terms, underscores the importance of saf...

Google's Affectionate Email-Replying AI Incident: A Case Study in AI Safety

Explore a recent incident involving Google's email-replying AI, discussing the importance of safe and secure AI, responsible AI governance,...

Exploring and Addressing Unintended Gender Bias in Google Image Search: A Step Towards Responsible AI

This analysis sheds light on an unintended gender bias in Google Image Search results, highlighting the importance of trustworthy AI and har...

Investigation into Critical Amazon Warehouse Bear Spray Incident: Emphasizing Responsible AI in Workplace Safety

This incident involving a critical bear spray accident at an Amazon warehouse serves as a poignant reminder of the importance of safety meas...

Exploring Potential Risks in Google's Ad Targeting System: A Responsible AI Perspective

Investigating an incident involving Google's ad-targeting system, this article highlights the importance of safe and secure AI. This AI inci...

Tesla on Autopilot Involved in Accident While Driver Watched Movie: A Reminder of the Need for Responsible AI Governance

This AI incident involving a Tesla vehicle on Autopilot highlights the importance of safe and secure AI governance. As this incident maps to...

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.