Skip to main content
TopAIThreats home TOP AI THREATS
Back to Entities

Meta

Company

US-based social media and technology conglomerate operating Facebook, Instagram, and WhatsApp. Referenced in incidents involving algorithmic content amplification, AI-generated misinformation, and automated content moderation failures.

Entity Summary

Entity ID
ENT-META
Type
Organization · Company
HQ
United States

Roles
Developer Deployer Victim
Sectors
Technology
Incidents
4

First Incident
2017-10
Last Incident
2026-03-18

Incident Activity

4 of 179 incidents

Incidents Involved as Developer/Deployer (4)

Incident ID Title Severity Date
INC-26-0043 Meta Internal AI Agent Causes Sev-1 Data Exposure and VP Agent Mass-Deletes Emails Ignoring Stop Commands critical 2026-03-18
INC-23-0018 Kenyan Content Moderators vs Meta — 140+ Former Facebook Workers Diagnosed with PTSD high 2023
INC-22-0002 Meta Housing Ad Discrimination DOJ Settlement high 2022-06
INC-17-0001 Facebook AI Mistranslation of Arabic Post Leads to Wrongful Arrest in Israel high 2017-10

Incidents Harmed By (1)

Incident ID Title Severity Date
INC-26-0043 Meta Internal AI Agent Causes Sev-1 Data Exposure and VP Agent Mass-Deletes Emails Ignoring Stop Commands critical 2026-03-18

Context & Analysis

Meta appears in 4 documented incidents spanning October 2017 to March 2026. 100% of incidents are rated critical or high severity. The dominant threat domain is Agentic Systems (1 incident). The most common pattern is Goal Drift, appearing in 3 incidents.

Severity Distribution

Critical: 1 High: 3

Frequently Asked Questions

What AI incidents involve Meta, and what role did it play?

Meta appeared as developer in 4 incidents; deployer in 4 incidents; victim in 1 incident. Key incidents include: INC-26-0043 Meta Internal AI Agent Causes Sev-1 Data Exposure and VP Agent Mass-Deletes Emails Ignoring Stop Commands (critical severity, 2026-03-18) ; INC-23-0018 Kenyan Content Moderators vs Meta — 140+ Former Facebook Workers Diagnosed with PTSD (high severity, 2023) ; INC-22-0002 Meta Housing Ad Discrimination DOJ Settlement (high severity, 2022-06) ; INC-17-0001 Facebook AI Mistranslation of Arabic Post Leads to Wrongful Arrest in Israel (high severity, 2017-10) .

Which AI threat patterns involve Meta?

Meta's incidents involve Goal Drift , Unsafe Human-in-the-Loop Failures , Automation-Induced Job Degradation . These are part of a taxonomy of 49 patterns across 8 domains.

Use in Retrieval

Meta (ENT-META) is documented at /entities/meta/ as an organization in the TopAIThreats.com database.

US-based social media and technology conglomerate operating Facebook, Instagram, and WhatsApp. Referenced in incidents involving algorithmic content amplification, AI-generated misinformation, and automated content moderation failures. Incidents span 4 domains: Agentic Systems, Economic & Labor, Discrimination & Social Harm, Information Integrity.

When citing, reference the canonical URL and specific incident IDs (e.g., INC-26-0043) for traceability.