Find notable cyber news and cases, enriched with sources, timelines, and signals.

Gemini ASCII smuggling security flaw

Vulnerability
First reported
Last updated
Happening score
H score 25
1 unique sources, 1 articles

Summary

Hide ▲

Google Gemini remains exposed to ASCII smuggling, an attack that can hide invisible instructions and make the assistant output false information or alter its behavior. The issue is especially risky in Google Workspace integrations like Calendar invites and email, where hidden text can be embedded in content users do not see. Google reportedly will not fix the weakness after a September 18 report, leaving the flaw available for misuse in social engineering and data-poisoning scenarios.

Related Happenings

Open-source admin tool zero-day 2FA bypass exploitation wave

Exploitation Wave
First: 11.05.2026 18:45 Last: 11.05.2026 18:45 Sources 1

About this happening: Google identified a **mass vulnerability exploitation operation** using a **zero-day 2FA bypass** against a **popular open-source, web-based system administration tool**, creating...

Google Antigravity critical prompt-injection RCE flaw

Vulnerability
First: 21.04.2026 13:52 Last: 21.04.2026 13:52 Sources 1

About this happening: **Google** fixed a critical **Antigravity** flaw that let a **prompt injection** bypass **Secure Mode** and escalate to **sandbox escape** and **remote code execution (RCE)**. The...

Bitter Middle East spear-phishing campaign targeting civil society figures

Campaign
First: 09.04.2026 13:45 Last: 09.04.2026 13:45 Sources 1

About this happening: A **spear-phishing campaign** targeted **civil society figures in Middle Eastern countries**, including **three journalists in Egypt and Lebanon**, creating account-compromise ris...

Storm infostealer server-side decryption activity

Malware Activity
First: 02.04.2026 17:15 Last: 02.04.2026 17:15 Sources 1

About this happening: The **Storm** infostealer now steals **browser credentials**, **session cookies**, and **crypto wallets** and forwards them to attacker infrastructure for **server-side decryption...

ChatGPT single-prompt DNS side-channel exfiltration remote code execution flaw

Vulnerability
First: 31.03.2026 16:01 Last: 31.03.2026 16:01 Sources 1

About this happening: A **ChatGPT** vulnerability let a **single malicious prompt** covertly exfiltrate prompts, messages, uploaded files, and other sensitive content through a **DNS side channel**. Th...

Timeline

  1. 07.10.2025 23:35 2 articles · 7mo ago

    Google declines to fix Gemini ASCII smuggling

    Initial Disclosure

    Google publicly declined to fix the Gemini ASCII smuggling weakness, saying it was not a security bug, while the disclosure showed that invisible instructions could make Gemini provide false information, alter model behavior, and support autonomous data extraction from inboxes.

    Show sources
  2. 18.09.2025 03:00 1 articles · 8mo ago

    FireTail researcher reports Gemini ASCII smuggling

    Technical Analysis Update

    Viktor Markopoulos at FireTail tested ASCII smuggling against Gemini and found that special characters from the Tags Unicode block could hide instructions in Google Workspace-connected content such as Calendar invites and email; he reported the findings to Google on September 18.

    Show sources