AI Sec Watch

OpenAI announced GPT-5.4-Cyber, a new AI model designed specifically for cybersecurity professionals, along with a three-part strategy to manage risks as AI becomes more powerful. The announcement comes after competitor Anthropic released a more limited version of its Claude Mythos model, citing concerns that advanced AI could be exploited by attackers, though OpenAI argues that current safeguards are sufficient for broad deployment of today's models.

CVE-2026-23653 is a command injection vulnerability (a flaw where an attacker can insert malicious commands into input that gets executed) in GitHub Copilot and Visual Studio Code that allows an authorized attacker to disclose information over a network. The vulnerability stems from improper neutralization of special elements used in commands. The CVSS severity score (a standard 0-10 rating of how serious a security flaw is) has not yet been assigned by NIST.

Anthropic confirmed it briefed the Trump administration about its new Mythos model, an AI system so dangerous it won't be released publicly due to powerful cybersecurity capabilities. The company is engaging with the government on AI safety issues while simultaneously suing the Department of Defense over a supply-chain risk label and disagreement over military access to Anthropic's systems.

Recent physical attacks targeting AI industry leaders, including an alleged Molotov cocktail attack on OpenAI CEO Sam Altman's home and gunfire at an official who supported a data center project, have raised concerns about safety in the AI industry. These incidents appear connected to activist concerns about AI's risks, including extinction fears and opposition to infrastructure expansion.

This academic survey examines how well large language model-based agents (AI systems that use LLMs to make decisions and take actions) can generalize, meaning how effectively they perform on new tasks or situations they weren't specifically trained for. The paper reviews research across different domains to understand what factors help or limit an agent's ability to adapt and work reliably in unfamiliar contexts.

Google is adding a new feature to Chrome called 'Skills' that lets you save your favorite Gemini prompts (instructions you give to AI) and reuse them across different webpages with a single click, instead of typing the same prompt repeatedly. This saves time when you want to perform the same AI task, like asking for vegan recipe substitutions, on multiple pages.

Kiro IDE (a development environment that uses AI agents to help developers) has a cross-site scripting vulnerability (XSS, where an attacker injects malicious code that runs in a web browser) in versions before 0.8.140. An attacker can exploit this by creating a malicious workspace with a crafted color theme name, and if a user opens and trusts that workspace, the attacker's code will execute on their computer.

Anthropic's new Mythos model is an AI designed for cybersecurity that can identify and exploit technical vulnerabilities better than most humans, but European regulators have been largely denied early access to it. The company limited initial access through Project Glasswing to a few US tech companies like Apple, Microsoft, and Amazon for security reasons, while most EU countries were excluded. European officials worry that private companies controlling access to such powerful technology raises concerns about national security and who should have influence over these systems.

A vulnerability in OpenAI Codex CLI v0.23.0 and earlier allows attackers to execute arbitrary code by creating malicious configuration files (.env and .codex/config.toml) in a repository. When a user runs the codex command in a compromised repository, the tool automatically loads these files without asking for permission, triggering the attacker's embedded commands.