Today's PHP Security Release, and How We're Strengthening PHP Security (PHP Foundation)
The PHP Foundation details why critical OOM vulnerabilities took significant time to resolve, citing the difficult trade-offs between backward compatibility, new denial-of-service risks, and the impact on third-party extensions. It highlights the Foundation's role in coordinating complex security fixes that individual maintainers might struggle to push through.
LWN · 23 October 2026 · Read the original →
[$] SUSE's efforts to disable setuid system-wide
SUSE's security team is proposing the total removal of setuid and setgid binaries from the distribution to eliminate a major class of privilege escalation vulnerabilities. The initiative involves reviewing every existing package and finding alternative permission models to maintain functionality without the inherent risks of setuid.
LWN · 23 October 2026 · Read the original →
[$] Hash-based module integrity for reproducible distributions
This post explores the technical conflict between ensuring kernel module integrity and maintaining reproducible builds. It examines a patch set that uses hash-based checking to provide security guarantees while allowing third parties to independently verify that the binaries match the source code.
LWN · 23 October 2026 · Read the original →
Opening New Paths for Open Source
The OSI is engaging with the UN and other international bodies to establish a common definition for 'Open Source AI.' This effort aims to ensure that policy and regulation are based on shared language, preventing fragmented legal requirements for AI developers and users.
Open Source Initiative · 23 October 2026 · Read the original →
Shared standards, independent review: Fin's AI lesson
A principal engineer at Fin explains how AI coding agents have shifted the engineering bottleneck from implementation to architectural governance. The lesson is that while code generation is now nearly free, the responsibility for system integrity and long-term maintainability has become more critical for senior leadership.
Tidelift · 23 October 2026 · Read the original →
Acting on the AI Shift in Cyber Risk With Prevention at the Codebase Level
The Five Eyes intelligence agencies have issued a warning that AI is shrinking the time between the discovery of a vulnerability and its active exploitation. This shift requires engineering leaders to move toward codebase-level prevention and automated verification rather than relying on reactive patching.
Tidelift · 23 October 2026 · Read the original →
GitHub Automates Open Source License Compliance
GitHub Enterprise Cloud now offers generally available ruleset-based checks to enforce centralized open source license policies for dependencies. This feature integrates license information into the dependency graph and Software Bill of Materials (SBOM) exports, and emits audit log events for policy changes.
github.blog · 23 October 2026 · Read the original →
LiteLLM PyPI/Docker Supply Chain Attack
On October 21, 2026, LiteLLM's release pipeline was compromised, leading to the publication of malicious v1.107.0 to PyPI and Docker. This version contained credential-stealing malware and remote code execution tools, requiring users who installed it to isolate systems and rotate credentials.
bleepingcomputer.com · 23 October 2026 · Read the original →
We are getting much faster at discovering how little we actually control.
8 stories, every Friday
Published here every week. Follow by RSS to get it as it lands.