Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

AI-accelerated attacks are redefining the threat landscape, but many of them still rely on one of the oldest tactics in the book: exploiting known vulnerabilities. The difference today is speed. Vulnerabilities that once took skilled hackers months or weeks to exploit can now be weaponized in hours or minutes. This acceleration is forcing organizations to rethink how they identify and remediate risk.

Six new security issues for every one issue remediated. That's the ratio Snyk research has found, and it's why the AI Security Engineers Community gave an hour of livestream time to fixing rather than finding. Play Video: Remediation Agents Demystified: Your AI Teammate for Fixing Security Bugs Remediation Agents Demystified paired a fireside chat with a live demo.

Blame for all flaws belongs to the flawed human author. Historically, the bottleneck for finding security bugs in software was human bandwidth. As pointed out in this great post by Tom Ptacek, it appears that large language models are exceptionally good at finding them with simple prompting. This adds substantial bandwidth to the effort of finding bugs.

An unauthenticated SQL injection vulnerability exists in the jsonArrayContains filter function in GeoTools, the geospatial library that underpins GeoServer. When an OGC filter is evaluated against a PostGIS data store, the value passed to jsonArrayContains is written into the generated SQL statement without escaping, allowing an attacker to alter the query the database executes.

CVE-2026-19478 is a code injection vulnerability in GitLab Community Edition and Enterprise Edition, caused by improper handling of a GraphQL directive in the application’s API layer. GitLab disclosed the issue and shipped fixes in an ad hoc critical patch release on August 17, 2026, outside its normal twice-monthly security schedule. The vulnerability carries a CVSS base score of 9.4 (Critical) as assigned by GitLab. Exploitation requires no authentication and no user interaction.

We benchmarked how well leading models produce vulnerability fixes that are both secure and functional, across ~150 real vulnerable code samples in JavaScript, Java, and Python. We ran each model on its own and with Snyk Intelligence (the new agentic Agent Fix architecture). The headline findings.

This post is an unfortunate affirmation of our prior research into abandoned open-source packages, where we found that 11% of the most-downloaded packages have been abandoned and not actively maintained, becoming invisible vulnerabilities to your scanner. Today we share a zip-slip vulnerability we found in extract-zip (CVE-2026-19693), an npm package with over 20 million weekly downloads.