CERT-FR: Multiple Spring AI Vulnerabilities, SQL Injection Risks

CERT-FR Security Advisories

Published March 17th, 2026

Detected March 17th, 2026

Summary

CERT-FR has issued an advisory regarding multiple vulnerabilities in Spring AI, versions 1.0.x prior to 1.0.4 and 1.1.x prior to 1.1.3. These vulnerabilities allow for SQL injection and security policy bypass. Users are advised to consult the vendor's security bulletins for patch information.

View original document View source feed page

What changed

CERT-FR has released an advisory (CERTFR-2026-AVI-0305) detailing multiple critical vulnerabilities discovered in Spring AI software. The identified issues, tracked as CVE-2026-22729 and CVE-2026-22730, affect Spring AI versions 1.0.x prior to 1.0.4 and 1.1.x prior to 1.1.3. Exploitation of these vulnerabilities could lead to SQL injection attacks and allow attackers to bypass security policies.

Organizations utilizing affected versions of Spring AI must immediately consult the provided security bulletins from Spring and apply the necessary patches or updates. Failure to address these vulnerabilities could expose systems to data breaches and unauthorized access. While no specific compliance deadline is mentioned, prompt remediation is critical to mitigate security risks.

What to do next

Review Spring AI version usage and identify affected installations (1.0.x < 1.0.4, 1.1.x < 1.1.3).
Consult Spring security bulletins for CVE-2026-22729 and CVE-2026-22730.
Apply vendor-provided patches or updates to mitigate SQL injection and security policy bypass risks.

Source document (simplified)

Premier Ministre S.G.D.S.N

Agence nationale
de la sécurité des
systèmes d'information

Paris, le 17 mars 2026 N° CERTFR-2026-AVI-0305 Affaire suivie par: CERT-FR

Avis du CERT-FR

Objet: Multiples vulnérabilités dans Spring AI

Gestion du document

Risques

Contournement de la politique de sécurité
Injection SQL (SQLi)

Systèmes affectés

Spring versions 1.0.x antérieures à 1.0.4
Spring versions 1.1.x antérieures à 1.1.3

Résumé

De multiples vulnérabilités ont été découvertes dans Spring AI. Elles permettent à un attaquant de provoquer une injection SQL (SQLi) et un contournement de la politique de sécurité.