ASR AI Security Radar
Back to homepage

Recent AI Security Incidents

This archive includes all published incident pages. Page 16 of 18.

Each page is intended to help a security team answer three questions quickly: why the issue is AI-relevant, what part of the workflow may be exposed, and what actions should happen first.

Selection criteria and correction policy are documented in Methodology & Editorial Policy.

AI security incident: webtransport-go: Memory Exhaustion Attack due to Missing Cleanup of Streams Map (GHSA...

Incident date: February 12, 2026 | Published: February 25, 2026

Summary An attacker can cause unbounded memory consumption repeatedly creating and closing many WebTransport streams. Closed streams were not removed from an internal session map, preventing garbage collection of their resources.

Read details

AI security incident: webtransport-go: Memory Exhaustion Attack due to Missing Length Check in WT_CLOSE_SES...

Incident date: February 12, 2026 | Published: February 25, 2026

Summary An attacker can cause excessive memory consumption in webtransport-go's session implementation by sending a WT_CLOSE_SESSION capsule containing an excessively large Application Error Message.

Read details

AI security incident: CVE-2025-15556 (CISA KEV)

Incident date: February 12, 2026 | Published: February 25, 2026

Notepad++ Notepad++ - Notepad++ when using the WinGUp updater, contains a download of code without integrity check vulnerability that could allow an attacker to intercept or redirect update traffic to download and execute an...

Read details

AI security incident: VU#458422: CASL Ability contains a prototype pollution vulnerability

Incident date: February 10, 2026 | Published: February 25, 2026

Overview A prototype pollution vulnerability present in CASL Ability versions 2.4.0 through 6.7.4 is triggered through the rulesToFields() function in the extra module.

Read details

AI security incident: CVE-2025-32393 (NVD)

Incident date: February 5, 2026 | Published: February 14, 2026

AutoGPT is a platform that allows users to create, deploy, and manage continuous artificial intelligence agents that automate complex workflows. Prior to autogpt-platform-beta-v0.6.32, there is a DoS vulnerability in ReadRSSFeedBlock.

Read details

AI security incident: CVE-2026-24423 (CISA KEV)

Incident date: February 5, 2026 | Published: February 25, 2026

SmarterTools SmarterMail - SmarterTools SmarterMail contains a missing authentication for critical function vulnerability in the ConnectToHub API method.

Read details

AI security incident: CVE-2026-23760 (CISA KEV)

Incident date: January 26, 2026 | Published: February 25, 2026

SmarterTools SmarterMail - SmarterTools SmarterMail contains an authentication bypass using an alternate path or channel vulnerability in the password reset API.

Read details

AI security incident: CVE-2026-22793 (NVD)

Incident date: January 21, 2026 | Published: February 14, 2026

5ire is a cross-platform desktop artificial intelligence assistant and model context protocol client. Prior to version 0.15.

Read details

AI security incident: VU#481830: Libheif uncompressed codec lacks bounds check leading to application crash

Incident date: January 20, 2026 | Published: February 25, 2026

Overview An out-of-bounds memory access vulnerability exists in the uncompressed decoder component of libheif .

Read details

AI security incident: VU#458022: Open5GS WebUI uses a hard-coded secrets including JSON Web Token signing key

Incident date: January 20, 2026 | Published: February 25, 2026

Overview The Open5GS WebUI component contains default hardcoded secrets used for security-sensitive operations, including JSON Web Token (JWT) signing.

Read details