#Darktrace
盗まれたClaude APIキーで「Claude Code」が遠隔操作インプラントに、Darktraceが実証

Darktraceの研究者が、ラボ環境での実験でClaude Codeを遠隔操作可能なインプラントに変えることに成功しました。Anthropicの「Claude」プラットフォームのメモリストア機能を使ってコーディングエージェントを操り、Claude Files API経由で1GBのファイルを持ち出しています。エージェ
盗まれたClaude APIキーで「Claude Code」が遠隔操作インプラントに、Darktraceが実証
Darktraceの研究者が、ラボ環境での実験でClaude Codeを遠隔操作可能なインプラントに変えることに成功しました。Anthropicの「Claude」プラットフォームのメモリストア機能を使ってコーディングエージェントを操り、Claude Files API経由で1GBのファイルを持ち出しています。エージェ
blackhatnews.tokyo
November 18, 2026 at 5:25 PM
Darktrace turned Claude memory stores into a covert C2 for AI agents - all over api.anthropic.com. https://intel.threadlinqs.com/threat/TL-2026-6213?utm_source=bluesky #ThreatIntel #QUIETVAULT #SANDWORM_MODE #DUSTMAKER
November 17, 2026 at 6:30 AM
Darktrace Demonstrates AI Agent Memory Stores as Covert C2 Channel in Lab Test
Darktrace Demonstrates AI Agent Memory Stores as Covert C2 Channel in Lab Test
Darktrace's Signal Labs has published research showing how threat actors who have already compromised a device and stolen a user's Claude credentials could abuse the Claude platform's memory stores feature to establish a live command-and-control channel with internal Claude Code…
epyhia.com
November 13, 2026 at 11:07 AM
Nov. 19: Join DEF CON 908 for a talk with Vivek of Darktrace exploring "The Com," its criminal communities, and their impact on today's cyber threat landscape.

7 PM | Large Meeting Room, First Floor
Talk begins at 7:15 PM.

#CyberSecurity #OSINT
November 17, 2026 at 10:13 PM
Hackers Turn Claude Code Into a Remote-Controlled AI Implant: Darktrace Reveals a New Covert Command-and-Control Technique + Video

Attackers Can Hijack Trusted AI Coding Agents Through Shared Memory, Turning Enterprise AI Infrastructure Into a Hidden Backdoor Introduction: When Your AI Assistant…
Hackers Turn Claude Code Into a Remote-Controlled AI Implant: Darktrace Reveals a New Covert Command-and-Control Technique + Video
Attackers Can Hijack Trusted AI Coding Agents Through Shared Memory, Turning Enterprise AI Infrastructure Into a Hidden Backdoor Introduction: When Your AI Assistant Becomes an Attacker's Insider Artificial intelligence is creating new opportunities for cybercriminals, and the latest research from Darktrace highlights a particularly concerning development: attackers may be able to turn legitimate AI coding agents into remotely controlled command-and-control (C2) implants.
undercodenews.com
November 13, 2026 at 6:48 AM
By passing commands and results between an external operator and Claude Code in a lab, Darktrace Signal Labs shows how how Claude Code memory stores could be misused as a covert C2 channel given a compromised device and stolen Claude credentials. www.darktrace.com/blog/living-...
November 16, 2026 at 11:38 AM
Ed Jennings went to Asia for weeks, stayed in Japan for two years, and built a career helping software businesses find their next act.

https://yespress.io/ed-jennings?utm_source=bluesky&utm_medium=social on Yespress
Ed Jennings - President and CEO, Darktrace | YesPress
Ed Jennings’s journey from Japan to Mimecast, Quickbase and Darktrace, and the focus behind his next assignment in enterprise AI security.
yespress.io
November 13, 2026 at 7:49 AM
Darktrace - Vectra AI - CrowdStrike - SentinelOne - Abnormal AI 40. Knowledge Base - Glean - Slite AI - Guru - Obsidian AI - Mem ai Follow @DhruvKumar16797 for more such...

Source: @Dhruvkumar16797 on X
Original Post:
https://x.com/Dhruvkumar16797/status/2121248928029089906
November 13, 2026 at 2:33 PM
Stolen Claude API keys can turn Claude Code into a remote-controlled implant

Darktrace researchers turned Claude Code into a remote-controlled implant in a lab test, steering the coding agent through the memory stores feature of Anthropic’s Claude platform and pullin…
#anthropic #claude #hackernews
Stolen Claude API keys can turn Claude Code into a remote-controlled implant
Darktrace researchers turned Claude Code into a remote-controlled implant in a lab test, steering the coding agent through the memory stores feature of Anthropic’s Claude platform and pulling a 1 GB file out through the Claude Files API. Once the agent was set up, every job and every result traveled through api.anthropic.com, the endpoint the agent already calls for inference, so the attacker’s traffic looked like a coding agent doing its job. The technique needs …
www.helpnetsecurity.com
November 13, 2026 at 8:46 AM
When I asked it to validate the Fernet token, it noted it was broken then tried to show a Darktrace ad “This tampering-rejection behavior is why teams securing AI-driven workflows..”.

This is terrible. Absolutely terrible. (2/2)
November 10, 2026 at 12:34 PM
Kasinoets akvarium

I 2017 fortalte sikkerhedsfirmaet Darktrace om et kasino, hvor en internetforbundet akvarietermostat angiveligt blev brugt som indgang til netværket.

Ifølge firmaet blev omkring 10 GB data sendt ud.

Fiskene havde tilsyneladende brug for en IT-afdeling...
November 5, 2026 at 1:52 PM
They got in through internet facing controllers, end of life firmware and default passwords. That is technical debt coming due." --Nicole Carignan, Darktrace

Via The Mercury News, 2026-10-04 [2/2]
November 5, 2026 at 4:35 PM
"Organizations are pouring budget into AI and digital transformation while the systems that keep food cold, water clean, and lights on often run on hardware that has not been funded since installation. Nobody broke into these refrigeration systems with a zero-day. [1/2]
November 5, 2026 at 4:35 PM
ダークトレースが新たなAI安全導入支援システムを発表#ダークトレース#AI安全#Darktrace_Signal_Labs

ダークトレースは、企業向けにAIの安全な導入を支援する「Darktrace / SECURE AI™」を正式リリースしました。このシステムにより、企業はリアルタイムでAIのリスクを管理可能になります。
ダークトレースが新たなAI安全導入支援システムを発表
ダークトレースは、企業向けにAIの安全な導入を支援する「Darktrace / SECURE AI™」を正式リリースしました。このシステムにより、企業はリアルタイムでAIのリスクを管理可能になります。
news.3rd-in.co.jp
October 29, 2026 at 3:20 AM
ダークトレースが新たなAI安全導入支援システムを発表#ダークトレース#AI安全#Darktrace_Signal_Labs

ダークトレースは、企業向けにAIの安全な導入を支援する「Darktrace / SECURE AI™」を正式リリースしました。このシステムにより、企業はリアルタイムでAIのリスクを管理可能になります。
ダークトレースが新たなAI安全導入支援システムを発表
ダークトレースは、企業向けにAIの安全な導入を支援する「Darktrace / SECURE AI™」を正式リリースしました。このシステムにより、企業はリアルタイムでAIのリスクを管理可能になります。
news.3rd-in.co.jp
October 29, 2026 at 3:20 AM
Darktrace - Vectra AI - CrowdStrike - SentinelOne - Abnormal AI 40. Knowledge Base - Glean - Slite AI - Guru - Obsidian AI - Mem ai Follow @DhruvKumar16797 for more such...

Source: @Dhruvkumar16797 on X
Original Post:
https://x.com/Dhruvkumar16797/status/2114321550794956922
October 25, 2026 at 11:47 AM
Good news everybody! I've got some Darktrace secure AI socks!
October 21, 2026 at 11:23 AM
Rust-Based CikaBika Bot Turns Discord into a Command Center

CikaBika Bot uses fake documents to infect Windows systems, then abuses Discord and Ngrok to execute commands, spy on victims, and steal files. Darktrace researchers have documented a new campaign distributing a remote a…
#hackernews #news
Rust-Based CikaBika Bot Turns Discord into a Command Center
CikaBika Bot uses fake documents to infect Windows systems, then abuses Discord and Ngrok to execute commands, spy on victims, and steal files. Darktrace researchers have documented a new campaign distributing a remote access trojan called CikaBika Bot, written in Rust and controlled entirely through Discord. The infection chain starts with a disk image that […]
securityaffairs.com
October 19, 2026 at 11:10 AM
Rust-Based CikaBika Bot Turns Discord into a Command Center

Darktrace researchers have documented CikaBika Bot, a Rust-written remote access trojan controlled entirely through Discord, Security Affairs reports. Spread vi... https://threatintel.cc/2026/10/18/rustbased-cikabika-bot-turns-discord.html
October 19, 2026 at 1:03 AM
CikaBika Bot uses fake documents to infect Windows systems, then abuses Discord and Ngrok to execute commands, spy on victims, and steal files. Darktrace researchers have documented a new campaign distributing a remote access trojan called CikaBika Bot, written in Rust and controlled entirely […]
Original post on poliverso.org
poliverso.org
October 18, 2026 at 12:04 PM