OpenAI

OpenAI has introduced GPT-Red, an internal automated red-teaming model designed to identify and remediate prompt injection vulnerabilities in GPT-5.6. This approach aims to tackle a growing safety challenge. While human red-team exercises are valuable, they cannot generate adversarial test cases...

A newly disclosed vulnerability in the OpenAI Codex desktop application for macOS could allow attackers to exploit indirect prompt injection techniques to exfiltrate sensitive data, according to a recent entry in the GitHub Advisory Database. Tracked as CVE-2026-14898, the issue...

Hackers are actively abusing OpenAI’s organization invitation feature to launch a new form of “poisoned tenant” attack, allowing them to harvest sensitive prompts, API activity, and potentially corporate data from unsuspecting users. According to research disclosed by Push Security, attackers...

OpenAI has introduced Daybreak, a strategic initiative to change how modern software is built and defended against emerging threats. Moving away from traditional reactive patching, Daybreak focuses on making software resilient by design from the very beginning of the development...

OpenAI has announced a new Bio Bug Bounty program for GPT-5.5 as part of its efforts to improve safety controls for advanced AI systems and to address misuse in biology. The initiative invites qualified researchers to test whether GPT-5.5 can...

OpenAI has officially launched the expanded phase of its Trusted Access for Cyber program. Granting select organizations access to its specialized GPT-5.4-Cyber model to strengthen digital defenses across critical infrastructure, financial services, and open-source security communities. The program operates on a tiered trust model...

The integration of AI coding agents has introduced new, high-impact attack surfaces for development teams. Phantom Labs at BeyondTrust recently discovered a critical command-injection vulnerability in OpenAI Codex. This flaw allowed attackers to steal sensitive GitHub User Access Tokens. By exploiting...

OpenAI has officially launched GPT-5.4 mini and GPT-5.4 nano, releasing its most capable small models designed to handle high-volume, latency-sensitive workloads. The new mini iteration offers a significant performance upgrade over the previous GPT-5 mini across reasoning, coding, tool use,...

OpenAI has announced its acquisition of Promptfoo, an artificial intelligence security platform designed to help enterprises find and fix vulnerabilities during development. This strategic move aims to secure AI systems against emerging threats, such as prompt injection and jailbreaks, before...

OpenClaw founder Peter Steinberger has officially joined OpenAI, marking a notable collaboration between open-source innovation and one of the world’s leading AI research organizations. According to Steinberger’s announcement titled “OpenClaw, OpenAI and the Future,” his new role focuses on advancing AI agents to make them accessible...