Incident 380: Facebook's Auto-Generated Targeting Ad Categories Contained Anti-Semitic Options

Description: Facebook's automated advertising categories generated using users' declared interests contained anti-Semitic categories such as "Jew hater" and "How to burn Jews" which were listed as fields of study.


New ReportNew ReportNew ResponseNew ResponseDiscoverDiscoverView HistoryView History
Alleged: Facebook developed and deployed an AI system, which harmed Jewish people.

Incident Stats

Incident ID
Report Count
Incident Date
Khoa Lam

Incident Reports

Reports Timeline

Facebook Enabled Advertisers to Reach ‘Jew Haters’ · 2017

Want to market Nazi memorabilia, or recruit marchers for a far-right rally? Facebook’s self-service ad-buying platform had the right audience for you. 

Until this week, when we asked Facebook about it, the world’s largest social network ena…

Updates to Our Ad Targeting · 2017

Facebook equips businesses with powerful ways to reach the right people with the right message. But there are restrictions on how audience targeting can be used on Facebook. Hate speech and discriminatory advertising have no place on our pl…

Facebook allowed advertisers to target 'Jew haters' · 2017

Facebook has allowed advertisers to target users interested in the topics of “Jew hater” and “How to burn Jews”, according to an investigation that adds to mounting criticisms of the way the company allows and profits from unethical ads.


Facebook's generation of 'Jew Hater' and other advertising categories prompts system inspection · 2017

Update: Facebook has disabled using the user-reported fields in question in its advertising system until further notice.

Update 2: Google has disabled similar functionality in its own ad targeting system as well. (Details below)

Facebook au…

Could Facebook Have Caught Its 'Jew Hater' Ad Targeting? · 2017

Facebook lives and dies by its algorithms. They decide the order of posts in your News Feed, the ads you see when you open the app, and which which news topics are trending. Algorithms make its vast platform possible, and Facebook can often…


A "variant" is an incident that shares the same causative factors, produces similar harms, and involves the same intelligent systems as a known AI incident. Rather than index variants as entirely separate incidents, we list variations of incidents under the first similar incident submitted to the database. Unlike other submission types to the incident database, variants are not required to have reporting in evidence external to the Incident Database. Learn more from the research paper.