'AgentCorruption' Puts AWS Environments At Risk With Single Prompt

Dark ReadingPublished 2026-10-09 14:58 UTC

A now-patched vulnerability in AWS Bedrock AgentCore could have allowed an attacker to use one AI chatbot to take over an organization's entire fleet.

— Dark Reading

Read at Dark Reading

Three fired OpenAI safety researchers dispute the company’s misconduct claims

Three former OpenAI safety researchers dispute the company’s claim that they mishandled sensitive information, saying they were dismissed for prioritising safety. In an open letter, they warn the firings may have a chilling effect on employees and AI safety culture.

56 minutes ago

Image: The Hacker News
Credential-Stealing GitHub Actions Workflows Planted in Tens of Thousands of Repositories
The Hacker NewsCybersecurity researchers have disclosed details of an ongoing credential-theft campaign that has compromised two high-profile open-source maintainer accounts to push a malicious workflow into over 340 repositories. "Using the account of Takashi Kitao, author of the 18,400-star…
2 hours ago
Image: The Hacker News
Attackers exploit unpatched AhsayCBS flaws to deploy webshells and cryptocurrency miners
Attackers are exploiting vulnerabilities in the AhsayCBS backup platform, deploying webshells and cryptocurrency miners. Sources describe the flaws as unpatched; one reports that they enable authentication bypass and OS command injection.
4 hours ago