Follow the latest coverage, related explainers and connected technology stories.
Two recent studies by the Microsoft Research team and the AI Red Team reveal how poisoned datasets and models can be exploited to plant backdoors in coding agents and execute malicious code remotely.
OpenAI has added persistent cloud development environments to Codex that can be accessed from different devices, along with a voice-enabled CLI interface and tools for reviewing code, scanning GitHub repositories, and automatically preparing security fixes.