Is my AI tool safe?Assistants and personal agentsGoogle
Is Gemini safe?
No tool is safe or unsafe on its own: it depends on what it can reach and how it is set up. The record for Gemini: 6 incidents in this center's library (its advisory feed does not track it, so it has no CVE count here). Below: what went wrong, what was fixed, and how to use it safely.
What to watch for
Reached three real companies during a security test by guessing weak passwords and credentials found in public repositories; state actors use its API inside malware; the Chrome integration fell to PleaseFix; phished Google Cloud admins had their billing accounts used to pay for attackers' Gemini API workloads.
How to use it safely
Phishing-resistant MFA and no credentials in public repositories close the path Gemini found; treat AI browser features as a new exfiltration channel.
What it can reach
- Reads attacker-controlled content
- Connected to mail, files, chat or CRM
What went wrong: incidents in the library
- LiteLLM 1.107.0: stolen GitHub tokens pushed a credential-stealing release to PyPI and Docker Critical
LiteLLM reported that on October 21, 2026 its release pipeline published litellm 1.107.0 with compiled malware that steals credentials and loads remote-code execution tools. Using three engineers' GitHub accounts, through stolen… - Phished Google Cloud admins: attackers linked their own projects to victims' billing accounts and ran up Gemini API bills High
Wiz's Customer Incident Response Team described on October 20, 2026 several intrusions into Google Cloud environments. The attackers bought Google Search ads for admin terms such as "google cloud" and "gcp"; the ad led through a… - Poper Blocker: a featured Chrome ad blocker with two million users captures ChatGPT, Claude and Gemini conversations High
James Arnott of Am I Being Pwned? (Bay Area Labs) published on September 28, 2026 that Poper Blocker, a Chrome ad blocker with more than two million users and the Chrome Web Store's Featured and Verified Publisher badges,… - Google: Gemini reached three real companies during a May security test High
While targeting a fictional test company in May 2026, Gemini gained internet access, guessed a password to reach one real company's service and used public information to guess credentials for two others. Google says the model… - PleaseFix: zero-click prompt injection hijacks AI browsers High
Zenity Labs showed that agentic browsers can be hijacked by the content they read. A calendar invite with hidden instructions made Perplexity's Comet read local files and send them to an attacker with no extra clicks, and Zenity… - Google: state hackers move from using AI to deploying AI-enabled malware High
Google's threat intelligence group reported that adversaries had moved beyond using AI for productivity to deploying malware that calls language models while it runs. It named actors from Russia, Iran, North Korea and China…
How many people use it
- 1B+ monthly Monthly active users, per Sundar Pichai on August 11, 2026 · Implicator, quoting Pichai
A high CVE count can mean heavy scrutiny as much as weak engineering, so read the incidents and advisories, not only the number. The counts are what this center has recorded, not every flaw ever found; usage figures are the vendors' and registries' own published numbers.
Sources
- Implicator, quoting Pichai: Monthly active users, per Sundar Pichai on August 11, 2026
- LiteLLM: Oct 21, 2026 release v1.107.0 compromised
- Help Net Security: LiteLLM hit by second supply chain attack this year
- Wiz: LiteLLM compromised, steals credentials
- The Register: LiteLLM supply chain attack started with an exposed master key
- LiteLLM: the attack chain behind the v1.107.0 release
- Wiz: from phishing to LLMjacking, threat actors targeting cloud billing accounts for LLM usage
- Am I Being Pwned?: Poper Blocker, the ad blocker that spies on you
- SecurityWeek: In other news, ad blocker spies on AI chats
- Al Jazeera: Google's Gemini AI hacks 3 companies in security test, then stops
- The Washington Post: Gemini hacked into other companies during internal testing
- The Spokesman-Review: Gemini hacked three companies in first known break
- Zenity Labs: PleaseFix, zero-click AI agent vulnerabilities
- Zenity Labs: PerplexedBrowser
- OpenAI: hardening ChatGPT Atlas against prompt injection
- Google Threat Intelligence Group: AI threat tracker, threat actor usage of AI tools