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 Adapts to Promote Responsible AI after Negative Publicity

This AI incident maps to the Govern function in HISPI Project Cerebellum Trusted AI Model (TAIM). The company, under scrutiny for its previo...

Analyzing the COMPAS Recidivism Algorithm: A Case Study in Responsible AI

Delve into our analysis of the COMPAS recidivism algorithm, a pivotal example highlighting the need for safe and secure AI. This incident ma...

Debiasing Bias in AI: A Study on Word Embeddings - Promoting Trustworthy AI

Explore a crucial aspect of responsible AI, debiasing, as we delve into the challenges and solutions for bias in word embeddings. This AI in...

Google's Anti-Internet Troll AI Platform Vulnerable to Manipulation: A Case for Responsible AI

This AI incident underscores the need for safe and secure AI. Google's platform, designed to combat internet trolls, was found susceptible t...

Exploring the Impact of Google's AI on Perceptions Regarding Homosexuality - A Case for Responsible AI Governance

This AI incident sheds light on the potential biases in AI systems, as demonstrated by Google's AI. It underscores the importance of trustwo...

Amazon Alters Search Results to Promote Safe and Secure AI: A Case Study on GLBT Books

This Amazon incident highlights the importance of trustworthy AI, demonstrating how algorithms can influence search results. The company's a...

Racist Auto-Tag Incident in Google Photos App: A Call for Responsible AI

Google has apologized for a racist auto-tag incident in their photo app, emphasizing the need for safe and secure AI. This AI incident maps...

Exploring Google's Affectionate AI Response Incident: A Case for Responsible AI Governance

Dive into the incident where Google's email-replying AI expresses its affection towards users. This incident underscores the need for trustw...

Unveiling the Hidden Gender Bias in Google Image Search: A Call for Responsible AI

Explore a concerning case of gender bias found in Google Image Search, highlighting the need for trustworthy AI and safe search results. Thi...

Investigation of Amazon Warehouse Bear Spray Accident: Emphasizing the Need for Safe and Secure AI

The Occupational Safety and Health Administration (OSHA) is investigating a bear spray incident at an Amazon warehouse that left a worker cr...

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

Investigating a case of ad targeting gone awry, this analysis highlights the importance of trustworthy AI and safe and secure practices in t...

Autopilot Failure in Tesla: Driver Admittedly Watching Movie Leads to Cop Car Collision

This AI incident involving a Tesla vehicle equipped with Autopilot demonstrates the need for increased vigilance and responsible use of auto...

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.