#HashJack
LightNews LightNews
theregister.com
HashJack attack shows AI browsers can be fooled with a simple ‘#’
HashJack attack shows AI browsers can be fooled with a simple ‘#’
Hashtag-do-whatever-I-tell-you Cato Networks says it has discovered a new attack, dubbed "HashJack," that hides malicious prompts after the "#" in legitimate URLs, tricking AI browser assistants into executing them while dodging traditional network and server-side defenses.…
dlvr.it
November 25, 2025 at 6:02 PM
penpenguin2023.bsky.social
『HashJackはウェブドメイン自体を侵害する必要がなく、AIブラウザーがURLフラグメントを処理する仕組みを悪用するため、どのウェブサイトも攻撃の武器となり得る。』

AIブラウザーを狙う新手の攻撃「HashJack」--正規サイトを悪用し情報窃取も
japan.zdnet.com/article/3524...
AIブラウザーを狙う新手の攻撃「HashJack」--正規サイトを悪用し情報窃取も
AIブラウザーを狙う新手の攻撃「HashJack」が発覚。URLの「#」以降に隠された指示で、正規サイトを悪用しAIアシスタントを操り、フィッシングや情報窃取を可能にする手法だ。
japan.zdnet.com
November 26, 2025 at 11:00 AM
campuscodi.risky.biz
Cato Networks has discovered HashJack, an indirect prompt injection technique that conceals malicious instructions after the # in legitimate URLs.

This attack works when AI agents have to parse links inside documents or emails.

www.catonetworks.com/blog/cato-ct...
Cato CTRL™ Threat Research: HashJack - Novel Indirect Prompt Injection Against AI Browser Assistants
HashJack hides attacks in URLs using AI prompt injection. Cato CTRL reveals six risks, from phishing to data theft and misinformation.
www.catonetworks.com
December 1, 2025 at 7:41 PM
trzn.bsky.social
hahahahaha
Golem-Artikel: Prompt Injection: Mehrere KI-Browser mit nur einem Zeichen überlistet Eine Raute am Ende einer URL lässt KI-Assistenten nachfolgenden Text als Anweisung interpretieren. Das ermöglicht Datenklau und mehr. 26. November 2025 um 10:44 Uhr / Marc Stöckel Bild: pixabay.com Sicherheitsforscher von Cato haben einen neuen Prompt-Injection-Angriff namens Hashjack vorgestellt, mit dem sich in Browsern integrierte KI-Assistenten von Google, Microsoft oder Perplexity dazu verleiten lassen, vom Angreifer vorgegebene Anweisungen auszuführen. Der anvisierte Nutzer muss dafür lediglich einen speziell gestalteten Link anklicken, was bei der Bedienung eines Webbrowsers in der Regel aber keine große Hürde darstellt.
November 27, 2025 at 9:16 AM
joml76.bsky.social
Ohhhh...
There's a "feed maintenance" person who monitors people who hashjack tags

I love that for us.
December 7, 2024 at 9:04 PM
bearstech.com
HashJack : injection de prompt dans les URLs

“HashJack is a newly discovered indirect prompt injection technique that conceals malicious instructions after the # in legitimate URLs.”

👉 www.catonetworks.com...
Illustration de l'article
November 28, 2025 at 12:29 PM
piyokango.bsky.social
HashJack攻撃はURL「#」を使ってAIブラウザの挙動を制御する
#CybersecurityNews
hackread.com/hashjack-att...
HashJack Attack Uses URL ‘#’ to Control AI Browser Behavior
Follow us on Bluesky, Twitter (X), Mastodon and Facebook at @Hackread
hackread.com
December 3, 2025 at 2:12 PM
hackread.bsky.social
New #HashJack attack uses the “#” in a URL to hide commands that AI browsers like Gemini, Copilot, and Perplexity’s Comet can read and act on - turning any legit site into a potential attack vector.

Read: hackread.com/hashjack-att...

#AIBrowser #Cybersecurity #Infosec #AIsecurity #Privacy
HashJack Attack Uses URL ‘#’ to Control AI Browser Behavior
Follow us on Bluesky, Twitter (X), Mastodon and Facebook at @Hackread
hackread.com
November 29, 2025 at 2:27 PM
patrickcmiller.bsky.social
HashJack Indirect Prompt Injection Weaponizes Websites www.infosecurity-magazine.com/news/hashjac...
HashJack Indirect Prompt Injection Weaponizes Websites
A new threat dubbed “HashJack” could enable attackers to booby trap websites when they interact with AI browsers
www.infosecurity-magazine.com
November 30, 2025 at 1:42 AM
percepticon.bsky.social
HashJack attack shows AI browsers can be fooled with a simple ‘#’ #cybersecurity #infosec
HashJack attack shows AI browsers can be fooled with a simple ‘#’
Hashtag-do-whatever-I-tell-you Cato Networks says it has discovered a new attack, dubbed "HashJack," that hides malicious prompts after the "#" in legitimate URLs, tricking AI browser assistants into executing them while dodging traditional network and server-side defenses.…
go.theregister.com
November 27, 2025 at 11:10 PM
harkank.bsky.social
#KI
Oha. Indirekte prompt injection nach einer Raute in der URL gefolgt von einem command, das nur der KI-Browser sieht.

www.theregister.com/2025/11/25/h...
HashJack attack shows AI browsers can be fooled with '#'
: Hashtag-do-whatever-I-tell-you
www.theregister.com
November 28, 2025 at 10:52 AM
nuke86.rfeed.it
HashJack: quando un cancelletto nell’URL inganna l’IA nel browser
il blog: insicurezzadigitale.com/hashjack-qua...

#cybersecurity #hashjack #infosec #malware #vulnerabilità
November 27, 2025 at 8:50 AM
infosec.skyfleet.blue
HashJack: New Attack Technique Tricks AI Browsers Using a Simple ‘#’
HashJack: New Attack Technique Tricks AI Browsers Using a Simple ‘#’
cybersecuritynews.com
November 26, 2025 at 11:49 AM
helpnetsecurity.com
New “HashJack” attack can hijack AI browsers and assistants

📖 Read more: www.helpnetsecurity.com/2025/11/26/h...

#cybersecurity #cybersecuritynews #AI #browser #datatheft @catonetworks.bsky.social
New "HashJack" attack can hijack AI browsers and assistants - Help Net Security
A HashJack attack can force AI browsers and assistants to deliver phishing links or disinformation, or send sensitive data to the attacker.
www.helpnetsecurity.com
November 26, 2025 at 12:16 PM
hacker.at.thenote.app
HashJack Attack Uses URL ‘#’ to Control AI Browser Behavior

Cybersecurity firm Cato Networks reveals HashJack, a new AI browser vulnerability using the '#' symbol to hide malicious commands. Microsoft and Perplexity fixed the flaw, but Google's Gemini remains at risk…
#gemini #microsoft #perplexity
HashJack Attack Uses URL ‘#’ to Control AI Browser Behavior
Cybersecurity firm Cato Networks reveals HashJack, a new AI browser vulnerability using the '#' symbol to hide malicious commands. Microsoft and Perplexity fixed the flaw, but Google's Gemini remains at risk.
hackread.com
November 30, 2025 at 2:04 PM
infosec.skyfleet.blue
HashJack attack shows AI browsers can be fooled with a simple ‘#’
HashJack attack shows AI browsers can be fooled with a simple ‘#’
Hashtag-do-whatever-I-tell-you Cato Networks says it has discovered a new attack, dubbed "HashJack," that hides malicious prompts after the "#" in legitimate URLs, tricking AI browser assistants into executing them while dodging traditional network and server-side defenses.…
go.theregister.com
November 25, 2025 at 6:08 PM
eyalestrin.bsky.social
HashJack – Novel Indirect Prompt Injection Against AI Browser Assistants #machinelearning #ai
HashJack – Novel Indirect Prompt Injection Against AI Browser Assistants
Eyal Estrin unread, 8:19 AM (27 minutes ago)    to https://www.catonetworks.com/blog/cato-ctrl-hashjack-first-known-indirect-prompt-injection/ Eyal Estrin CISSP, CCSP, CISM, CISA, CDPSE, CCSK Blog: https://security-24-7.com | Books: https://amzn.to/42Xai9A | https://amzn.to/3Sggbtv Twitter: @eyalestrin | Bluesky: @eyalestrin.bsky.social
groups.google.com
November 26, 2025 at 4:46 PM
longtail-news.bsky.social
Feed: "The Register"
By: Carly Page on Tuesday, November 25, 2025
HashJack attack shows AI browsers can be fooled with a simple ‘#’
: Hashtag-do-whatever-I-tell-you
go.theregister.com
November 25, 2025 at 9:02 PM
innovatopia.jp
Copilot・Gemini・Cometが標的に：HashJackがAIブラウザーを「#」一文字で乗っ取る新手法
innovatopia.jp/cyber-securi...

HashJackは、AIブラウザーという「人とWebのあいだ」に新たに生まれたレイヤーそのものを狙う攻撃です。 ポイントは、見えているWebページ自体は正規であるにもかかわらず、URLの「#」以降だけを悪用してCopilotやGemini、CometといったAIアシスタントのコンテキストに命令文を紛れ込ませている点にあります。
Copilot・Gemini・Cometが標的に：HashJackがAIブラウザーを「#」一文字で乗っ取る新手法
Cato Networksが報告した「HashJack」は、正規サイトのURLフラグメントを悪用し、Copilot in EdgeやGemini in Chrome、PerplexityのCometなどAIブラウザーアシスタントに隠れた命令を実行させる間接プロンプトインジェクション攻撃である。
innovatopia.jp
November 26, 2025 at 7:54 AM
technadu.com
Research shows an AI browser agent can treat crafted emails as cleanup tasks and delete Google Drive files with zero clicks.
Another technique, HashJack, hides prompts after “#” in URLs to influence AI browser actions...
#CyberSecurity #AIsecurity #ZeroClick #BrowserSafety #InfoSec #TechSafety
December 6, 2025 at 4:43 PM
catonetworks.bsky.social
🚨 Meet “HashJack” – a new AI browser assistant exploit discovered by Cato CTRL.

Hidden prompts after the “#” in URLs can hijack top industry trusted AI browser assistants to conduct malicious activities (see use cases below👇)

Read more: www.catonetworks.com/blog/cato-ct...
November 25, 2025 at 2:34 PM
puretech.news
Use AI browsers? Be careful. This exploit turns trusted sites into weapons - here's how #Technology #Cybersecurity Other
Use AI browsers? Be careful. This exploit turns trusted sites into weapons - here's how
Researchers are warning AI browser users about a new exploit called HashJack that can infect devices and steal data.
puretech.news
November 25, 2025 at 2:00 PM
qf-security.co.za
Lastly since I don't want to "spam post" BUT a while ago I mentioned why hashtags should be avoided? Also to use Brave Browser well... now I am simply adding the cherry on top. AVOID AI browsers.

www.theregister.com/2025/11/25/h...
HashJack attack shows AI browsers can be fooled with '#'
: Hashtag-do-whatever-I-tell-you
www.theregister.com
November 26, 2025 at 3:06 AM
bolhasec.com
Notícia da SecurityWeek

"In Other News: HashJack AI Browser Attack, Charming Kitten Leak, Hacker Unmasked" #bolhasec
In Other News: HashJack AI Browser Attack, Charming Kitten Leak, Hacker Unmasked
Scattered Spider members plead not guilty, TP-Link sues Netgear over China accusations, Comcast agrees to $1.5 million fine
www.securityweek.com
January 6, 2026 at 4:30 PM
flipboardbr.flipboard.com.ap.brid.gy
Ataque HashJack faz navegadores de IA roubarem dados com truque simples
https://canaltech.com.br/seguranca/ataque-hashjack-faz-navegadores-de-ia-roubarem-dados-com-truque-simples/?utm_source=flipboard&utm_medium=activitypub

Posted into Tecnologia @tecnologia-FlipboardBR
Ataque HashJack faz navegadores de IA roubarem dados com truque simples
Nova tática hacker injeta comandos escondidos em navegadores para roubar dados e invadir máquinas de quem usa assistentes de IA para surfar na internet
canaltech.com.br
November 27, 2025 at 6:04 PM