HTML Injection Vulnerability in ChatGPT by OpenAI
CVE-2025-43714

6.5MEDIUM

Key Information:

Vendor

OpenAI

Status
Vendor
CVE Published:
19 May 2025

Badges

📈 Score: 259👾 Exploit Exists📰 News Worthy

What is CVE-2025-43714?

CVE-2025-43714 is a vulnerability found in the ChatGPT system developed by OpenAI, which is a platform designed for natural language processing and conversational AI. This vulnerability stems from the way the system handles the rendering of Scalable Vector Graphics (SVG) documents. Instead of treating them as plain text within a secure environment, the ChatGPT interface allows for inline rendering, which can lead to HTML injection attacks when exploited. This flaw could allow malicious users to inject harmful HTML content, manipulating the interface and potentially executing unauthorized scripts within the web browsers of users interacting with the application.

Such a vulnerability could have dire consequences for organizations employing ChatGPT. If exploited, it could compromise user data, manipulate the AI's responses, or facilitate phishing attacks, enabling attackers to deceive users into divulging sensitive information.

Potential Impact of CVE-2025-43714

  1. Phishing Attacks: The vulnerability could be utilized to create convincing phishing interfaces that trick users into providing confidential information, such as login credentials or payment details, leading to significant security breaches.

  2. Data Manipulation and Loss: Attackers might exploit this flaw to manipulate the data presented to users or affect the integrity of communications, leading to misinformation or unauthorized access to sensitive data.

  3. Compromise of User Sessions: By injecting harmful scripts via HTML, malicious actors could seize control of user sessions, allowing them to impersonate legitimate users and execute unauthorized actions, potentially resulting in further exploitation of organizational resources.

News Articles

ChatGPT SVG Vulnerability: CVE-2025-43714

Discover the ChatGPT SVG flaw (CVE-2025-43714) and rising phishing threats. Learn how GrackerAI helps turn risks into content opportunities.

References

CVSS V3.1

Score:
6.5
Severity:
MEDIUM
Confidentiality:
Low
Integrity:
Low
Availability:
Low
Attack Vector:
Network
Attack Complexity:
Low
Privileges Required:
None
User Interaction:
None
Scope:
Unchanged

Timeline

  • 👾

    Exploit known to exist

  • 📰

    First article discovered by GrackerAI

  • Vulnerability published

  • Vulnerability Reserved

.
CVE-2025-43714 : HTML Injection Vulnerability in ChatGPT by OpenAI