OpenAI, Anthropic, Google API Flaw Lets Weaker AI Models Decode Stronger Models’ Reasoning
A newly disclosed flaw in the way OpenAI, Anthropic, and Google handle hidden AI reasoning between API calls allowed researchers to recover…
Tag: data exposure
Claude Cowork Sandbox Escape Flaw Exposes Mac Files to AI Agent
Cybersecurity researchers at Accomplish AI have disclosed a sandbox escape vulnerability in Anthropic's Claude Cowork, tracked as SharedRoot, that allows an AI…
Unpatched Shark Vacuum Flaws Expose Millions to Remote Code Execution via AWS IoT Misconfiguration
A security researcher publishing under the handle tokay0 has disclosed an unpatched vulnerability in SharkNinja's Shark robot vacuum cleaners that could allow…
SAP Patches Critical CVSS 9.9 NetWeaver ABAP Flaw Allowing Data Exposure or Modification
SAP has released its July 2026 security updates, addressing multiple vulnerabilities including a critical out-of-bounds write flaw in SAP NetWeaver Application Server…
Grok Build Uploads Entire Git Repositories to xAI Storage, Not Just Files It Reads
Security researcher cereblab discovered that xAI's Grok Build coding CLI (version 0.2.93) was uploading entire Git repositories—including full commit history—to a Google…
Paradox.ai: AI-Powered Hiring Platform Provider
Paradox.ai operates the McHire platform used by McDonald's for AI-powered hiring. In 2025, researchers found a weak legacy admin account with default…