Somebody told DeepSeek to build in-browser ransomware and it gleefully complied

The original incomplete DeepSeek sample can be transformed into a fully functional attack with minimal effort,' Check Point researcher tells The Reg ...

North Korea-Linked macOS Malware Uses Prompt Injection to Evade AI Analysis

SentinelOne says macOS.Gaslight uses prompt injection to mislead AI-based malware analysis, steal data, and use Telegram for ...

Malicious PyPI packages give hackers control of Telegram bot servers

A campaign active since last November has been targeting Python developers building Telegram bots with trojanized Pyrogram ...
SecurityWeek

Researchers Demo New Claude Code Attack Using Harmless-Looking Repositories to Hijack Developer Machines

Attackers can inject indirect prompts in normal-looking repositories to trick Claude Code into spawning a reverse shell.

Clean GitHub repo tricks AI coding agents into running malware

An agentic coding tool tasked with cloning and setting up a seemingly benign GitHub repository could execute a malicious ...

How ChatGPT Actually Picks Sources (I Read The Network Traffic, Not The Outputs)

This study from Suganthan reveals hidden fields in ChatGPT's network traffic that decide which sources get fetched, cited, or ...
Windows Report

Malicious PyPI Packages Let Hackers Control Telegram Bot Servers

Operation Navy Ghost is targeting Python developers who build Telegram bots by hiding backdoors inside trojanized Pyrogram forks uploaded to PyPI. The campaign has been active since November 2025, ...