Researchers escaped OpenAI's Codex sandbox two ways, one running commands on a developer's machine from its most locked-down mode. OpenAI has patched both.

Two OpenAI Codex sandbox flaws, Overpatch and Heapjack, could let malicious repositories execute commands on developer systems.

AM JST today, includes four combined fixes for paths where credentials that should have been masked were appearing in logs ...

Optimization of over 300 TiB of memory and migration of 800,000 lines of code to RustUpdated: 2026/10/03Executive ...

Prism's allowlist is not ours and it changes without notice: gpt-6-astra was accepted in the morning of 2026-09-17 and rejected by the afternoon with 400: Unsupported assistant model. When that ...

Unsloth details how Studio scans model code, blocks flagged weights, inspects packages and sandboxes tools before anything ...

In other news: South Korea raises the breach fine level; three Turkish ministers targeted with spyware; CISA is ready to hire 250 staff. AI company Anthropic has disclosed a fourth incident where one ...

Google on Tuesday released updates to patch 230 security vulnerabilities, including one that has come under active exploitation in the wild. The medium-severity vulnerability, assigned the CVE ...

Google has rolled out Chrome 153 to the stable channel for Windows, Mac, and Linux, shipping as version 153.0.8010.36 on Linux and 153.0.8010.36/.37 on Windows and Mac.