Embed Notice

HTML Code

<blockquote style="position: relative; padding-left: 55px;"><section><a href="https://chaos.social/users/evawolfangel/statuses/115546888209252616">Eva Wolfangel (evawolfangel@chaos.social)'s status on Friday, 14-Nov-2025 17:31:48 JST</a><a href="https://chaos.social/@evawolfangel" title="evawolfangel@chaos.social"><img src="https://gnusocial.jp/avatar/8922-48-20240602170211.webp" width="48" height="48" alt="Eva Wolfangel" style="position: absolute; left: 0; top: 0;">Eva Wolfangel</a></section><article><p>Ich sags ungern, aber: Told you so.<br>Vor genau diesem Szenario warne ich seit Monaten in diversen Vorträgen, Podcasts, Texten. Oft kommt zurück: "Ach komm, jetzt mach doch nicht immer alles schlecht. Wir können nicht auf KI verzichten, es machen doch alle..." Und: "Diese Attacken gibt es doch nur in der Theorie."<br>Voila: Hier eine reale Attacke des chinesischen Geheimdienstes mit der Hilfe von KI Agenten von Anthropic. <br><a href="https://www.anthropic.com/news/disrupting-AI-espionage" rel="nofollow">https://www.anthropic.com/news/disrupting-AI-espionage</a></p><p><a href="https://chaos.social/tags/aiagents" rel="tag">#aiagents</a> <a href="https://chaos.social/tags/anthropic" rel="tag">#anthropic</a> <a href="https://chaos.social/tags/chatgpt" rel="tag">#chatgpt</a> <a href="https://chaos.social/tags/cybersecurity" rel="tag">#cybersecurity</a></p></article><footer><a rel="bookmark" href="https://gnusocial.jp/conversation/5861570#notice-11514517">In conversation</a><time datetime="2025-11-14T17:31:48+09:00" title="Friday, 14-Nov-2025 17:31:48 JST">about 9 months ago</time> <span>from <span><a href="https://chaos.social/@evawolfangel/115546888209252616" rel="external" title="Sent from chaos.social via ActivityPub">chaos.social</a></span></span><a href="https://chaos.social/@evawolfangel/115546888209252616">permalink</a><h4>Attachments</h4><ol><li><label><a rel="external" href="https://gnusocial.jp/attachment/5454631">Screenshot aus dem Bericht:They also told
Claude that it was an employee of a
legitimate cybersecurity firm, and
was being used in defensive testing.
The attackers then initiated the
second phase of the attack, which
involved Claude Code inspecting the
target organization’s systems and
infrastructure and spotting the
highest value databases. Claude was
able to perform this reconnaissance
in a fraction of the time it would've
taken a team of human hackers. It
then reported back to the human
operators with a summary of its
findings.In the next phases of the attack,
Claude identified and tested security
vulnerabilities in the target
organizations’ systems by
researching and writing its own
exploit codes Having done so, the
framework was able to use Claude to</a></label><br><a href="https://assets.chaos.social/media_attachments/files/115/546/868/558/474/988/original/35996affc3b10322.jpg" rel="external">https://assets.chaos.social/media_attachments/files/115/546/868/558/474/988/original/35996affc3b10322.jpg</a></li><li><article><header><div>Domain not in remote thumbnail source whitelist: www.anthropic.com</div><h5><a href="https://www.anthropic.com/news/disrupting-AI-espionage#aiagents">Disrupting the first reported AI-orchestrated cyber espionage campaign</a></h5><div> from <span>@AnthropicAI</span></div></header><div>A report describing an a highly sophisticated AI-led cyberattack</div><footer></footer></article></li></ol></footer></blockquote>

Corresponding Notice

Embed this notice
Eva Wolfangel (evawolfangel@chaos.social)'s status on Friday, 14-Nov-2025 17:31:48 JST Eva Wolfangel
Ich sags ungern, aber: Told you so.
Vor genau diesem Szenario warne ich seit Monaten in diversen Vorträgen, Podcasts, Texten. Oft kommt zurück: "Ach komm, jetzt mach doch nicht immer alles schlecht. Wir können nicht auf KI verzichten, es machen doch alle..." Und: "Diese Attacken gibt es doch nur in der Theorie."
Voila: Hier eine reale Attacke des chinesischen Geheimdienstes mit der Hilfe von KI Agenten von Anthropic.
https://www.anthropic.com/news/disrupting-AI-espionage
#aiagents #anthropic #chatgpt #cybersecurity
In conversationabout 9 months ago from chaos.socialpermalink
Attachments
1. Screenshot aus dem Bericht: They also told Claude that it was an employee of a legitimate cybersecurity firm, and was being used in defensive testing. The attackers then initiated the second phase of the attack, which involved Claude Code inspecting the target organization’s systems and infrastructure and spotting the highest value databases. Claude was able to perform this reconnaissance in a fraction of the time it would've taken a team of human hackers. It then reported back to the human operators with a summary of its findings. In the next phases of the attack, Claude identified and tested security vulnerabilities in the target organizations’ systems by researching and writing its own exploit codes Having done so, the framework was able to use Claude to
  https://assets.chaos.social/media_attachments/files/115/546/868/558/474/988/original/35996affc3b10322.jpg
2. Domain not in remote thumbnail source whitelist: www.anthropic.com
  Disrupting the first reported AI-orchestrated cyber espionage campaign
  from @AnthropicAI
  A report describing an a highly sophisticated AI-led cyberattack

Public

Embed Notice

HTML Code

Corresponding Notice