A seemingly harmless PNG image could fool AI coding assistants into exposing sensitive data, according to new security ...

A PNG hiding a prompt injection could steal your repo's secrets, researchers demonstrate. The technique, dubbed 'Ghostcommit, ...

IBM Bob multi-agent AI platform gains Bobalytics cost analytics and isolated subagents on July 9, one day after Microsoft ...

Awareness of all the ways prompt injection can be effected will help security teams spot a new generation of attacks.

A reverse shell gives an attacker interactive command-line access to a compromised machine by making the target reach out first. Instead of the attacker connecting inward to a listening port on the ...

Attackers have begun backdooring internet-exposed Ivanti Sentry appliances, the nonprofit security watchdog Shadowserver confirmed on June 11, 2026 — less than 48 hours after patches and a public ...

Google has fixed a critical flaw in its agentic integrated developer environment (IDE) Antigravity that led to sandbox escape and remote code execution (RCE) after researchers created a proof of ...

Cybersecurity researchers have discovered a critical "by design" weakness in the Model Context Protocol's (MCP) architecture that could pave the way for remote code execution and have a cascading ...

A design flaw – or expected behavior based on a bad design choice, depending on who is telling the story – baked into Anthropic's official Model Context Protocol (MCP) puts as many as 200,000 servers ...

In the high-velocity era of AI-driven development, the "engineering productivity paradox" has revealed a sobering truth: while tools can now generate code at a blistering pace, they often create a ...

Enterprise software maker SAP on Tuesday announced the release of 17 new security notes as part of its January 2026 Security Patch Day. Four of the notes address critical-severity vulnerabilities. The ...