Malicious Prompts Could Exploit ChatGPT to Steal Data From Your Private Emails, Claims Researcher

Malicious prompts could exploit ChatGPT to steal data from your private emails, claims researcher

An X user demonstrated a security flaw in ChatGPT’s new MCP tools on X, showing that a malicious calendar invite could trick the AI into leaking emails and calendar data. Though developer mode requires approval, users may unknowingly expose sensitive information.

Compromised Google Calendar invites can hijack ChatGPT’s Gmail connector and leak emails

A security researcher has demonstrated how a malicious Google Calendar invite can prompt-inject ChatGPT and coax it into leaking private emails once Google connectors are enabled.

Vitalik Warns of AI Governance Risks as ChatGPT Security Flaw Surfaces

TLDR: Vitalik warns AI governance models can be exploited with jailbreak prompts and malicious inputs. ChatGPT MCP tools enabled email access, exposing private data to attackers via calendar invites. Info finance model proposed to ensure AI model diversity and human oversight in governance. AI agents can be tricked using simple methods, highlighting security risks for crypto users. Ethereum co-founder Vitalik Buterin has sounded the alarm on AI…

ChatGPT leaks private email data, Vitalik Buterin is pissed

Open the article to view the coverage from Cryptopolitan

ChatGPT’s New Support For MCP Tools Let Attackers Exfiltrate All Private Details From Email - Cybernoz - Cybersecurity News

A newly introduced feature in ChatGPT that allows it to connect with personal data applications can be exploited by attackers to exfiltrate private information from a user’s email account. The attack requires only the victim’s email address and leverages a malicious calendar invitation to hijack the AI agent. On Wednesday, OpenAI announced that ChatGPT would begin supporting Model Context Protocol (MCP) tools, an innovation from AnthropicAI desi…