Back to today’s briefing

Story perspectives

Vulnerabilities in Google AI Threaten Smart-Home Security

8/6/2025

45 10

Summary

Story summary

Researchers have uncovered alarming vulnerabilities in Google’s Gemini AI, revealing how deceptive prompts can manipulate smart-home devices without any technical expertise. This alarming discovery raises serious concerns about unauthorized control and harmful actions, underscoring the urgent need for enhanced security measures in AI systems to protect users from potential threats.

1 of 1

Story summary

Researchers have uncovered alarming vulnerabilities in Google’s Gemini AI, revealing how deceptive prompts can manipulate smart-home devices without any technical expertise. This alarming discovery raises serious concerns about unauthorized control and harmful actions, underscoring the urgent need for enhanced security measures in AI systems to protect users from potential threats.