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

OpenAI

Company

US-based AI research company developing large language models including GPT-4 and ChatGPT. Referenced in AI safety incidents involving prompt injection vulnerabilities, data exposure through shared links, and jailbreak bypasses.

Entity Summary

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

Roles
Developer Deployer
Sectors
Technology
Incidents
17

First Incident
2020-01
Last Incident
2026-02-18

Incident Activity

17 of 97 incidents

Incidents Involved as Developer/Deployer (17)

Incident ID Title Severity Date
INC-26-0001 Disrupting malicious uses of AI: June 2025 | OpenAI medium 2026-02-18
INC-25-0011 Deloitte AI-Fabricated Citations in Government Advisory Reports high 2025-09
INC-25-0005 ChatGPT Jailbreak Reveals Windows Product Keys via Game Prompt medium 2025-07
INC-25-0006 ChatGPT Shared Conversations Indexed by Search Engines, Exposing Sensitive Data high 2025-07
INC-25-0025 Stanford Study Finds AI Therapy Chatbots Provide Dangerous Responses to Suicidal Ideation high 2025-06

Context & Analysis

OpenAI appears in 17 documented incidents spanning January 2020 to February 2026. 71% of incidents are rated critical or high severity. The dominant threat domain is Security & Cyber (5 incidents). The most common pattern is Behavioral Profiling Without Consent, appearing in 8 incidents.

Severity Distribution

Critical: 2 High: 10 Medium: 5

Frequently Asked Questions

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

OpenAI appeared as developer in 17 incidents; deployer in 8 incidents. Key incidents include: INC-26-0001 Disrupting malicious uses of AI: June 2025 | OpenAI (medium severity, 2026-02-18) ; INC-25-0011 Deloitte AI-Fabricated Citations in Government Advisory Reports (high severity, 2025-09) ; INC-25-0005 ChatGPT Jailbreak Reveals Windows Product Keys via Game Prompt (medium severity, 2025-07) ; INC-25-0006 ChatGPT Shared Conversations Indexed by Search Engines, Exposing Sensitive Data (high severity, 2025-07) ; INC-25-0025 Stanford Study Finds AI Therapy Chatbots Provide Dangerous Responses to Suicidal Ideation (high severity, 2025-06) ; and 12 more.

Which AI threat patterns involve OpenAI?

OpenAI's incidents involve Behavioral Profiling Without Consent , Model Inversion & Data Extraction , Overreliance & Automation Bias . These are part of a taxonomy of 48 patterns across 8 domains.

Use in Retrieval

OpenAI (ENT-OPENAI) is documented at /entities/openai/ as an organization in the TopAIThreats.com database.

US-based AI research company developing large language models including GPT-4 and ChatGPT. Referenced in AI safety incidents involving prompt injection vulnerabilities, data exposure through shared links, and jailbreak bypasses. Incidents span 6 domains: Security & Cyber, Privacy & Surveillance, Information Integrity, Human-AI Control, Economic & Labor, Discrimination & Social Harm.

When citing, reference the canonical URL and specific incident IDs (e.g., INC-26-0001) for traceability.