ChatGPT for Google Sheets Vulnerability Enables Data Exfiltration and Phishing Overlay Attacks (promptarmor.com)

• Indirect prompt injection in ChatGPT for Google Sheets allows data exfiltration across multiple workbooks
• Attack includes phishing overlay and attacker-controlled sidebar
• OpenAI removed Apps Script code generation after delayed response

"Security firm PromptArmor discovered that OpenAI's ChatGPT extension for Google Sheets is vulnerable to indirect prompt injection attacks, allowing an attacker to exfiltrate multiple workbooks from a victim's account and display phishing overlays. The attack works even when the user has disabled automatic edits. OpenAI responded by removing the model's ability to generate Apps Script code."