Skip to content
Submit incident
AII-1002-51919
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

ReDoS vulnerability in Giskard text perturbation detector

A Regular Expression Denial of Service (ReDoS) vulnerability was discovered in Giskard's text perturbation detector, affecting versions prior to 2.15.5. The vulnerability in a URL-detection regex could cause excessive computation times or system crashes when processing specific text patterns.

14 November 2024 Β· Giskard-AI Β· Security and cyberattacks
14 November 2024
AII-1002-519192 sources ↓
Quick facts
Category
Security and cyberattacks
Source
github.com
Date
14 November 2024
Drafted by
Raidu Incident Detection

What happened

A Regular Expression Denial of Service (ReDoS) vulnerability was discovered in Giskard's text perturbation detector by the GitHub Security Lab team. The vulnerability exists in a regex used to detect URLs and links in Giskard's punctuation removal transformation, which is susceptible to catastrophic backtracking when processing specific text patterns.

Reported impact

The vulnerability could trigger exponential regex evaluation times, potentially causing extended computation times or crashes in Giskard when processing text containing certain patterns.

Index analysis

Giskard versions prior to 2.15.5 are affected. The vulnerability was assigned CVE-2024-52524 and classified as Low severity. A fix was released in version 2.15.5.

Governance controls

The vulnerability was discovered by GHSL team member Kevin Backhouse and reported through GitHub's security advisory process.

Classification

Category
Security and cyberattacks
Industry
Digital security
Harm status
Potential harm
Intent
Unintentional
Timing
Unknown
AI systems
Giskard Scan
Regulations
FCRA (cited)

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

github.com
https://github.com/Giskard-AI/giskard/security/advisories/GHSA-pjwm-cr36-mwv3
cve.org
NVD entry