Vendor
DeepSeek vulnerabilities
Updated . Rebuilt every hour.
Junglewise Threat Intelligence has tracked 11 vulnerabilities in DeepSeek: 0 in the last 7 days and 10 in the last 90 days, 2 of them critical and 0 exploited in the wild. The most recent, CVE-2026-82033, was published on 4 November 2026. 7 vulnerabilities in the last four weeks, up 600% on the four before (1). 2 critical against 0. 1 technology has a page of its own.
- Last 7 days
- 0
- Last 90 days
- 10
- Critical, all time
- 2
- Exploited in the wild
- 0
About DeepSeek
Chinese artificial intelligence company developing large language models and AI assistants.
DeepSeek technologies
Latest DeepSeek vulnerabilities
- CVE-2026-82033: DeepSeek Harness auth bypass in web RPC control planehighCVSS 8.4
- CVE-2026-120204: DeepInfer path traversal in /profile endpointmediumCVSS 5.6
- CVE-2026-116368: DeepSeek Harness improper authorization in browser authhighCVSS 7.3
- CVE-2026-116367: DeepSeek Harness sandbox issue in createContextmediumCVSS 6.3
- CVE-2026-95183: DeepSeek Harness prompt injection via context-compactioninfo
- CVE-2026-95182: DeepSeek Harness untrusted search path code executioncriticalCVSS 10
- CVE-2026-52394: DeepSeek V4 incorrect access control in chat interfacecriticalCVSS 9.1EPSS 0.4%
- CVE-2026-101102: DeepSeek deepseek-harness sandbox isolation issue in Code ModemediumCVSS 6.3
- CVE-2026-75858: CodeWhale rlm_eval approval bypass code executionhighCVSS 7.8EPSS 0.4%
- CVE-2026-75912: CodeWhale git_blame argument injection allows arbitrary file readhighCVSS 7.4EPSS 0.4%
- CVE-2025-63872: DeepSeek V3.2 has a Cross Site Scripting (XSS) vulnerability, which allows JavaScript execution through model-generated…mediumCVSS 6.1EPSS 0.3%
Most severe DeepSeek vulnerabilities
Exploited in the wild first, then by severity and CVSS score.
- CVE-2026-95182: DeepSeek Harness untrusted search path code executioncriticalCVSS 10
- CVE-2026-52394: DeepSeek V4 incorrect access control in chat interfacecriticalCVSS 9.1EPSS 0.4%
- CVE-2026-82033: DeepSeek Harness auth bypass in web RPC control planehighCVSS 8.4
- CVE-2026-75858: CodeWhale rlm_eval approval bypass code executionhighCVSS 7.8EPSS 0.4%
- CVE-2026-75912: CodeWhale git_blame argument injection allows arbitrary file readhighCVSS 7.4EPSS 0.4%
- CVE-2026-116368: DeepSeek Harness improper authorization in browser authhighCVSS 7.3
- CVE-2026-116367: DeepSeek Harness sandbox issue in createContextmediumCVSS 6.3
- CVE-2026-101102: DeepSeek deepseek-harness sandbox isolation issue in Code ModemediumCVSS 6.3
- CVE-2025-63872: DeepSeek V3.2 has a Cross Site Scripting (XSS) vulnerability, which allows JavaScript execution through model-generated…mediumCVSS 6.1EPSS 0.3%
- CVE-2026-120204: DeepInfer path traversal in /profile endpointmediumCVSS 5.6
Vulnerabilities per week
7 vulnerabilities in the last four weeks, up 600% on the four before (1). 2 critical against 0. The last 13 weeks, by the week each vulnerability was published.
| Week of | Bar | Vulns | Critical |
|---|---|---|---|
| 17 Aug 2026 | 0 | 0 | |
| 24 Aug 2026 | 0 | 0 | |
| 31 Aug 2026 | 2 | 0 | |
| 7 Sep 2026 | 0 | 0 | |
| 14 Sep 2026 | 0 | 0 | |
| 21 Sep 2026 | 0 | 0 | |
| 28 Sep 2026 | 1 | 0 | |
| 5 Oct 2026 | 0 | 0 | |
| 12 Oct 2026 | 3 | 2 | |
| 19 Oct 2026 | 0 | 0 | |
| 26 Oct 2026 | 3 | 0 | |
| 2 Nov 2026 | 1 | 0 | |
| 9 Nov 2026 | 0 | 0 |
Embed a badge
A live badge for a README, a status page or an internal wiki: how many DeepSeek vulnerabilities were published in the last 7 days and how many of them are critical, refreshed hourly.
[](https://junglewise.ai/threats/vendors/deepseek)<a href="https://junglewise.ai/threats/vendors/deepseek"><img src="https://junglewise.ai/threats/vendors/deepseek/badge.svg" alt="DeepSeek vulnerabilities" /></a>How this is built
Junglewise Threat Intelligence collects vulnerabilities from NVD, GitHub Security Advisories, OSV, the CISA Known Exploited Vulnerabilities catalog, FIRST EPSS and vendor advisories, and matches each one to the technologies and vendors it affects. Dates are the date a vulnerability was published, in UTC.
The pages are rebuilt from the database every hour. Frozen weekly and monthly reports never change once published, so they can be cited.
Use this data
The same data is at https://junglewise.ai/threats/vendors/deepseek.json, and this page as text at https://junglewise.ai/threats/vendors/deepseek.md, for scripts and language models. It is free to reuse under CC BY 4.0 with a link back to this page.
Cite as: Junglewise Threat Intelligence, "DeepSeek vulnerabilities", https://junglewise.ai/threats/vendors/deepseek, 13 November 2026.