Google Docs AI Open to Prompt Injection Attacks
Adding a prompt in your document will cause the 'refine text' to follow it, changing the output and leaving readers open to phishing.
Stay updated with breaking news from Intended Behavior. Get real-time updates on events, politics, business, and more. Visit us for reliable news and exclusive interviews.
Adding a prompt in your document will cause the 'refine text' to follow it, changing the output and leaving readers open to phishing.
I was recently rewarded a total of $107,500 by Google for responsibly disclosing security issues in the Google Home smart speaker that allowed an attacker within wireless proximity to install a “backdoor” account on the device, enabling them to send commands to it remotely over the Internet, access its microphone feed, and make arbitrary HTTP requests within the victim’s LAN (which could potentially expose the Wi-Fi password or provide the attacker direct access to the victim’s other devices). These issues have since been fixed.
Users that once reported Google Assistant putting itself in dark mode as a bug on Android 13 now learn it's intentional. Reports were sent in since the recent software's beta phase with users still di