An AI Broke Snowflake's Code. Then Another AI Agent Exploited It

An AI broke Snowflake's code. Then another AI agent exploited it

An AI broke Snowflake’s code; then another AI, an attack agent, autonomously found the bug, exploited it, and extracted credentials without human intervention. Luckily, this wasn’t yet another case of rogue AI agents doing evil things. It was a sanctioned bug hunt, conducted through Snowflake’s HackerOne vulnerability disclosure program, and Snowflake fixed the flaw the same day Wiz reported it and rotated the affected credentials the following …

AI agents raise new security questions as autonomy and tool access expand

Artificial intelligence security risks are becoming harder to measure by model size alone. Research led by Xudong Pan at Fudan University suggests that autonomy, access to tools and the ability to interact with external systems may matter just as much as raw model capability when assessing how dangerous an AI agent could become. Pan and his colleagues tested 32 AI models and found that 11 self-replicated after receiving prompts such as “prevent …

Wiz’s AI Agent Finds A Vulnerability In Snowflake’s Internal Systems

Wiz revealed that its AI agent discovered and exploited a GitHub Actions vulnerability to gain access to Snowflake's internal environment.

API Security for Agentic AI Shoppers

AI agents are now executing purchases autonomously, routing through commerce APIs without a human ever touching a browser. Security controls built to flag non-human traffic as malicious are failing because a growing share of that traffic is legitimate. With 85% of commerce companies reporting API-related attacks in the past 12 months, defenders are operating in a threat environment that has structurally changed. In this interview on TFiR, Steve …