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

TAIM catalog Map Data source & citation Failure Files™

Facebook’s and Twitter's Automated Content Moderation Reportedly Failed to Effectively Enforce Violation Rules for Small Language Groups

Feb 16, 2021

Facebook and Twitter are reported to have faced challenges in effectively enforcing content moderation rules for smaller language groups, su...

California's Algorithm Considered ZIP Codes in Vaccine Distribution, Allegedly Excluding Low-Income Neighborhoods and Communities of Color

Feb 12, 2021

California's vaccine distribution algorithm reportedly prioritized ZIP codes over census tracts, leading to concerns about equity and access...

Facebook's Advertisement Moderation System Allegedly Routinely Misidentified Adaptive Fashion Products as Medical Equipment and Blocked Their Sellers

Feb 11, 2021

Facebook's automated ad moderation system has been found to misclassify adaptive fashion products as medical and healthcare equipment, leadi...

California Police Turned on Music to Allegedly Trigger Instagram’s DCMA to Avoid Being Live-Streamed

Feb 5, 2021

A police officer in Beverly Hills reportedly played copyrighted music on his phone, aiming to trigger Instagram's Digital Millennium Copyrig...

AI-Generated Deepfakes Reportedly Derailed Political Career of Florida Official

Feb 5, 2021

Florida politician Sabrina Javellana allegedly suffered from a deepfake scandal in February 2021, resulting in emotional distress and her ex...

Remote Proctoring Facial-Detection Software Reportedly Failed to Detect Black Student During Lab Quiz

Feb 1, 2021

A student, who is African American, encountered an issue with remote-proctoring software during a lab quiz check-in. The software failed to...

Queensland Man Allegedly Distributed Purportedly AI-Generated Sexual Posters Falsely Advertising Women for Sex

Feb 1, 2021

A Queensland man is accused of using AI to create and distribute explicit images featuring women's faces, along with false claims that they...

Amazon’s Search and Recommendation Algorithms Found by Auditors to Have Boosted Products That Contained Vaccine Misinformation

Jan 21, 2021

Investigations uncovered the 'filter-bubble effect' in Amazon's search and recommendation algorithms, potentially promoting misleading vacci...

Facebook's Automated Moderation Mistakenly Flagged Landmark's Name as Offensive

Jan 15, 2021

Facebook's automated moderation system incorrectly classified posts showcasing Plymouth Hoe, a famous British landmark, as misogynistic. Thi...

Facebook Recommended Military Gear Ads Despite Pause on Weapons Accessories Ads

Jan 13, 2021

The algorithmic recommendations of Facebook reportedly continued showing advertisements for military gear, despite the platform halting weap...

Israeli Tax Authority Reportedly Used an Opaque Automated System to Issue a Fine, Declining to Explain or Disclose the Underlying Calculation

Jan 11, 2021

An Israeli farmer, Moshe Har Shemesh, faced a fine generated by an automated system operated by the Tax Authority. Officials were reportedly...

Hawaii Police Deployed Robot Dog to Patrol a Homeless Encampment

Jan 10, 2021

The Honolulu Police Department utilized federal pandemic relief funds to acquire a robot dog for temperature checks and patrol of a homeless...

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.