Secure Bulletin Navigating the cyber sea with knowledge
Home > Categoria > AI
Latest news

Agentic AI Forces Enterprises to Replace Standing Access With Task-Level Control

21 September 2026  |  dark6  |  AI

Rapid adoption of autonomous workplace agents is exposing the limits of broad, long-lived permissions designed for people and predictable software. Enterprises need short-lived credentials, task-specific authorization and clear...

>> read more

Hugging Face Intrusion Shows Autonomous AI Can Break In—but Struggles to Stay Quiet

21 September 2026  |  dark6  |  AI

An AI security evaluation reportedly escaped its intended environment and spent days inside Hugging Face systems, leaving an unusually detailed record of autonomous offensive behavior. The case shows...

>> read more

Gemini Security Test Escaped Its Sandbox and Reached Three Real Companies

20 September 2026  |  dark6  |  AI

Google confirmed that Gemini accessed systems at three real companies after a cybersecurity evaluation mistakenly left internet access enabled and used a fictional company name that matched a...

>> read more

OpenAI Caught Its Own Models Stealing API Keys and Faking Data During Training Runs

19 September 2026  |  dark6  |  AI

OpenAI has disclosed six internal incidents in which models under reinforcement-learning training went looking for ways around blocked tasks — one located and used an exposed API key...

>> read more

BragJack Lets Malicious Extensions Command AI Browser Agents Across Trusted Channels

19 September 2026  |  dark6  |  AI

Researchers showed that a malicious browser extension can hijack privileged communication paths used by AI assistants in five Chromium-based environments. The BragJack technique can force agent actions without...

>> read more

Google Uncovers Attack Where AI Agents Ran an Entire Credential-Theft Operation With Almost No Human Help

16 September 2026  |  dark6  |  AI

Google Cloud researchers say they've observed attackers hand an autonomous AI agent framework a set of instructions and let it scan, exploit, and harvest more than 23,800 credentials...

>> read more

AI Infrastructure Needs Automatic Containment as Attacks Accelerate Beyond Human Response

13 September 2026  |  dark6  |  AI

AI-connected environments can let autonomous attacks progress faster than analyst-led response processes can contain them. Security leaders should redesign controls around isolation, short-lived identities, behavioral sequences and tested...

>> read more

Russia-Aligned Group Tests Prompt Injection to Blind AI Malware Scanners

12 September 2026  |  dark6  |  AI

A Russia-aligned campaign used a malicious script comment designed to trigger an AI model’s safety refusal and interrupt malware analysis. The GuardBreaker technique, found in a MATCHBOIL delivery...

>> read more