Anthropic gives EU access to Mythos months after model’s release

Mythos is so powerful at uncovering cybersecurity flaws that the company limited access to a select group of vetted institutions to root out and patch vulnerabilities before bad actors exploit them under its Project Glasswing initiative. — AFP

Anthropic PBC has handed the European Union (EU)’s cybersecurity agency access to its powerful Mythos artificial intelligence (AI) model, more than three months after first signalling the bloc would be given access.

"Following our constructive engagement with Anthropic, we can confirm that the EU’s cybersecurity agency ENISA has been granted access to Mythos 5 and is testing it now,” Thomas Regnier, a spokesperson for the EU’s executive branch, the European Commission, said in an email.

Washington accuses Chinese AI firms of 'systematic' theft

Anthropic researcher resigns with warning about the dangers of AI development

Corporate America is getting hooked on open-source AI

Thank you for your report!

Anthropic reports fourth cybersecurity incident with early version of Claude

Thomson Reuters detects cybersecurity incident, says unauthorized party accessed files

By opening a model, a Chinese AI lab may test the world’s cybersecurity

Mythos在发现网络安全漏洞方面功能强大，因此该公司将其访问权限限制在经过严格审查的特定机构范围内，以便在恶意行为者利用漏洞之前将其根除并修复，这项计划名为“玻璃之翼计划”（Project Glasswing）。——法新社

Anthropic PBC 已向欧盟网络安全机构提供了其强大的 Mythos 人工智能 (AI) 模型的访问权限，此前该公司已于三个多月前首次表示将向欧盟提供访问权限。

欧盟执行机构欧盟委员会发言人托马斯·雷尼尔在一封电子邮件中表示：“经过与 Anthropic 的建设性接触，我们可以确认，欧盟网络安全局 ENISA 已获准访问 Mythos 5，目前正在对其进行测试。”

华盛顿指责中国人工智能公司“系统性”窃取数据

人类学研究员辞职，并警告人工智能发展带来的危险

美国企业界正沉迷于开源人工智能

感谢您的报告！

Anthropic公司报告了与早期版本的Claude相关的第四起网络安全事件。

汤森路透检测到网络安全事件，称未经授权方访问了文件

通过开放模型，中国人工智能实验室或许能够测试全球网络安全。