Description

Two high-severity security vulnerabilities have been made public in the popular image editing software GIMP that can allow remote attackers to execute arbitrary code. CVE-2025-2760 and CVE-2025-2761 are the names of these weaknesses, which affect all versions of GIMP prior to 3.0.0 and were made public on April 7th, 2025. Both vulnerabilities require user interaction, such as opening up malicious files or accessing malicious web sites, to be exploited. The discovery was made by security researcher Michael Randrianantenaina and it was submitted via the Zero Day Initiative (ZDI). The first vulnerability, CVE-2025-2760, is an integer overflow attack weakness in GIMP X Window Dump (XWD) file handling.Inadequate verification of user-provided input can lead to unsafe memory allocation when reading files. This creates a scenario where attackers can exploit to execute arbitrary code under the GIMP process with potential system compromise depending on user privileges. The second vulnerability, CVE-2025-2761, relates to how GIMP handles FLI (FLIC animation) files. It is a product of an out-of-bounds write caused by invalid file data verification, making it possible for attackers to write to memory and execute code. The flaw was first reported to the vendor on March 9th, 2025, and also permits remote code execution with the same privileges as the current user. Both bugs were resolved in GIMP 3.0.0, released on March 16th, 2025. The GIMP developers resolved the two bugs by implementing stronger input validation.The users are recommended to upgrade to version 3.0.0 or higher. Organisations should also provide user awareness training on risks of opening suspicious files. Vendors like SUSE and Amazon Linux have released security updates for their platforms with mitigation guidance.