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

DeepSeek

Company

Chinese AI company developing open-weight large language models. Referenced in incidents involving model security vulnerabilities and data governance concerns related to open-weight model distribution.

Entity Summary

Entity ID
ENT-DEEPSEEK
Type
Threat Actor · Company
Region
China

Roles
Developer Deployer Threat Actor
Sectors
Technology
Incidents
5

First Incident
2025
Last Incident
2026-02-09

Incident Activity

5 of 179 incidents

Incidents Involved as Developer/Deployer (5)

Incident ID Title Severity Date
INC-26-0025 Microsoft GRP-Obliteration: Single Prompt Reverses Safety Alignment Across 15 LLMs high 2026-02-09
INC-26-0083 DeepSeek Mass Government Bans and Publicly Exposed Database with 1M+ Records high 2026-01
INC-25-0029 Chain-of-Thought Reasoning Jailbreak Exploits Thinking Models high 2025-02
INC-25-0003 DeepSeek R1 Data Exposure and International Bans Over Privacy and Security Concerns high 2025-01
INC-25-0034 Chinese AI Labs Conduct Industrial-Scale Distillation Attacks Against Claude critical 2025

Incidents as Threat Actor (1)

Incident ID Title Severity Date
INC-25-0034 Chinese AI Labs Conduct Industrial-Scale Distillation Attacks Against Claude critical 2025

Context & Analysis

DeepSeek appears in 5 documented incidents spanning 2025 to February 2026. 100% of incidents are rated critical or high severity. The dominant threat domain is Security & Cyber (3 incidents). The most common pattern is Model Inversion & Data Extraction, appearing in 4 incidents.

Severity Distribution

Critical: 1 High: 4

Frequently Asked Questions

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

DeepSeek appeared as developer in 3 incidents; deployer in 4 incidents; threat actor in 1 incident. Key incidents include: INC-26-0025 Microsoft GRP-Obliteration: Single Prompt Reverses Safety Alignment Across 15 LLMs (high severity, 2026-02-09) ; INC-26-0083 DeepSeek Mass Government Bans and Publicly Exposed Database with 1M+ Records (high severity, 2026-01) ; INC-25-0029 Chain-of-Thought Reasoning Jailbreak Exploits Thinking Models (high severity, 2025-02) ; INC-25-0003 DeepSeek R1 Data Exposure and International Bans Over Privacy and Security Concerns (high severity, 2025-01) ; INC-25-0034 Chinese AI Labs Conduct Industrial-Scale Distillation Attacks Against Claude (critical severity, 2025) .

Which AI threat patterns involve DeepSeek?

DeepSeek's incidents involve Model Inversion & Data Extraction , Jailbreak & Guardrail Bypass , Mass Surveillance Amplification . These are part of a taxonomy of 49 patterns across 8 domains.

Use in Retrieval

DeepSeek (ENT-DEEPSEEK) is documented at /entities/deepseek/ as a threat actor in the TopAIThreats.com database.

Chinese AI company developing open-weight large language models. Referenced in incidents involving model security vulnerabilities and data governance concerns related to open-weight model distribution. Incidents span 2 domains: Security & Cyber, Privacy & Surveillance.

When citing, reference the canonical URL and specific incident IDs (e.g., INC-26-0025) for traceability.