Ars Technica

New hack uses prompt injection to corrupt Gemini’s long-term memory

blows my mind that businesses are even considering putting sensitive information into these leaky things. while this was an intended cracking of an attempt at a wall, clearly they leak info ...

AI-controlled robot shoots human after prompt manipulation, raising serious safety risks

A viral YouTube experiment has triggered fresh alarm across the AI and robotics community after an AI-controlled robot fired ...
CRN

Partners Say Alleged CIA Hacking Papers Prompt 'Constant Vigilance' In Mobile Security

Alleged CIA attempts to defeat smartphone security measures in order to obtain data from phones underscore how seriously companies must take their efforts around enterprise mobility security, solution ...
Washington Examiner

WH taking hack of U.S. Central Command ‘seriously’

The White House said Monday it was taking the hack of the U.S. Central Command Twitter feed “seriously” but cautioned that the incident was far different from the type of “large data breach” that has ...
Seeking Alpha

Varonis at Infosecurity Europe: Preventing Copilot Prompt-Hacking and Data Exposure

LONDON, May 29, 2024 (GLOBE NEWSWIRE) -- Varonis Systems, Inc. (VRNS) (Nasdaq: VRNS), a leader in data security, announced its full conference schedule for Infosecurity Europe, taking place June 4 – 6 ...
Ars Technica

New hack uses prompt injection to corrupt Gemini’s long-term memory

In the nascent field of AI hacking, indirect prompt injection has become a basic building block for inducing chatbots to exfiltrate sensitive data or perform other malicious actions. Developers of ...