Cyber Resilience

CVE-2026-4996

Medium

Published: 28 March 2026

Published
28 March 2026
Modified
24 April 2026
KEV Added
Patch
CVSS Score v4 6.9 CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
EPSS Score 0.0004 13.1th percentile
Risk Priority 14 60% EPSS · 20% KEV · 20% CVSS

Summary

CVE-2026-4996 is a medium-severity Injection (CWE-74) vulnerability. Its CVSS base score is 6.9 (Medium).

Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked at the 13.1th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as NLP and Transformers; in the Data-Related Vulnerabilities risk domain.

The strongest mitigations our analysis identified are NIST 800-53 SI-10 (Information Input Validation) and SI-2 (Flaw Remediation).

Deeper analysis

CVE-2026-4996 is a SQL injection vulnerability (CWE-74, CWE-89) in Sinaptik AI PandasAI versions up to 0.1.4. The issue affects the pandasai-lancedb extension, specifically the functions delete_question_and_answers, delete_docs, update_question_answer, update_docs, get_relevant_question_answers_by_id, and get_relevant_docs_by_id in the file extensions/ee/vectorstores/lancedb/pandasai_lancedb/lancedb.py.

The vulnerability enables remote exploitation with network access, low attack complexity, no privileges, no user interaction, and unchanged scope, as reflected in its CVSS v3.1 base score of 7.3 (AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L). Attackers can manipulate inputs to these functions, achieving low impacts on confidentiality, integrity, and availability.

VulDB advisories note that the vendor was contacted early for disclosure but provided no response, with no patches or mitigations detailed. A public exploit is available via a GitHub Gist, alongside VulDB entries documenting the issue.

This vulnerability targets an AI/ML-related component involving PandasAI and the LanceDB vector store extension, and the publicly available exploit heightens the risk of active use.

EU & UK References

Vulnerability details

A vulnerability was identified in Sinaptik AI PandasAI up to 0.1.4. Affected by this issue is the function delete_question_and_answers/delete_docs/update_question_answer/update_docs/get_relevant_question_answers_by_id/get_relevant_docs_by_id of the file extensions/ee/vectorstores/lancedb/pandasai_lancedb/lancedb.py of the component pandasai-lancedb Extension. Such manipulation leads to sql injection. The attack can be launched remotely.…

more

The exploit is publicly available and might be used. The vendor was contacted early about this disclosure but did not respond in any way.

CWE(s)

AI Security AnalysisAI

AI Category
NLP and Transformers
Risk Domain
Data-Related Vulnerabilities
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: ai, lancedb

Related Threats

MITRE ATT&CK Enterprise TechniquesAI

T1190 Exploit Public-Facing Application Initial Access
Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network.
Why these techniques?

SQL injection (CWE-89) in exposed pandasai-lancedb functions allows remote unauthenticated exploitation over the network (AV:N/PR:N), directly mapping to initial access via public-facing application exploitation. Low C/I/A impacts align with data manipulation in the vector store but do not enable additional techniques such as RCE or credential dumping.

Confidence: HIGH · MITRE ATT&CK Enterprise v18.1

CVEs Like This One

CVE-2026-3150Shared CWE-74, CWE-89
CVE-2026-3746Shared CWE-74, CWE-89
CVE-2025-2683Shared CWE-74, CWE-89
CVE-2026-5238Shared CWE-74, CWE-89
CVE-2026-4288Shared CWE-74, CWE-89
CVE-2026-2220Shared CWE-74, CWE-89
CVE-2025-1535Shared CWE-74, CWE-89
CVE-2026-0597Shared CWE-74, CWE-89
CVE-2026-1688Shared CWE-74, CWE-89
CVE-2026-5018Shared CWE-74, CWE-89

Affected Assets

Mitigating Controls

Mitigating Controls (NIST 800-53 r5) AI

prevent

Directly prevents SQL injection attacks like CVE-2026-4996 by validating and sanitizing inputs to vulnerable functions such as delete_question_and_answers before they reach LanceDB queries.

prevent

Ensures timely remediation of known flaws like this SQL injection vulnerability in PandasAI LanceDB extension through identification, reporting, testing, and installation of patches or workarounds.

preventdetect

Supports proactive discovery of SQL injection vulnerabilities such as CVE-2026-4996 via regular vulnerability scanning and monitoring of affected components.

References