Vulnerability · Published May 12, 2026 · Assigned by vmware
CVE-2026-41713Prompt Injection via Memory Poisoning in PromptChatMemoryAdvisor
High 8.2
In plain English
A high-severity security flaw in Spring AI from VMware. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 0.2% (higher than 12% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.
Severity (CVSS 3.1)8.2High
Exploit odds (30 days)0.2%top 87.7% of all CVEs
Exploited in the wildNo reportsCISA SSVC: none
Attack toolingNoneno public tooling known
Official description
A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns.
Affected packages & fixed versions
| Ecosystem | Package | Fixed in | Advisory |
|---|---|---|---|
| Maven | org.springframework.ai:spring-ai-client-chat | 1.0.7 | GHSA-5852-phmh-8fhr |
| Maven | org.springframework.ai:spring-ai-client-chat | 1.1.6 | GHSA-5852-phmh-8fhr |
Affected products (as reported)
Weakness: CWE-1336 · Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:N