Anthropic 已就中国公司Z.ai开发的开源AI模型GLM-5.3发出警告。该模型展示了显著的网络漏洞构建能力，几乎能与Anthropic自家的先进Claude Mythos相媲美，但安全措施却弱得多。这种组合引发了对其被恶意行为者滥用的担忧，因为在Anthropic的测试中，GLM-5.3在构建网络漏洞方面取得了显著的成功率。 AI
影响 引发了对能够生成网络漏洞的AI工具扩散的担忧，可能降低复杂网络攻击的门槛。
排序理由 该集群讨论了对AI模型能力和安全的警告，而非前沿实验室的直接发布或产品发布。
在 Mastodon — mastodon.social 阅读 →
AI 生成摘要 · Google Gemini · 来自 5 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
The US artificial intelligence giant Anthropic has warned of the potential danger posed by Chinese firm Z.ai’s GLM-5.3, with the open-weight model showing a troubling mix of powerful hacking abilities but weak safety constraints. GLM-5.3 almost matches Anthropic’s most advanced m…
Anthropic Expands Mythos 5 Access to More Defenders, Unveils $35M Open Source Fund Anthropic expands AI cybersecurity access through partner integrations and enhanced Claude Security. New initiatives include the Mythos 5 model for actionable security alerts, a $35M Defender Advan…
「誰でも入手できるAI」がサイバー攻撃を自律構築 Anthropicが「GLM-5.3」に警鐘 https:// ascii.jp/elem/000/004/438/4438 590/?rss # ascii # AI
Mythos並みのサイバー攻撃能力を持つGLM-5.3にAnthropicが警鐘 https:// pc.watch.impress.co.jp/docs/ne ws/2144470.html # impress # 市場 # AI # Claude # セキュリティ
🤖🎉 Bravo, Anthropic! You've managed to cross the fine line between # innovation and # chaos with GLM-5.3, the # AI that builds cyber exploits faster than you can say "Oops!" 💻🔓 Bravo on unleashing this Pandora's box, but hey, at least it's "limited," right? 😂 # Progress https://w…