Skip to content
The Nexus
DossierENTITY

UK's AI Security Institute

Coverage of UK's AI Security Institute in the Nexus archive.

Earliest in view: Jul 23 · 08:55 UTCMost recent: Aug 5 · 12:47 UTC
Co-mentioned in this coverage
Anthropicorganization3OpenAIorganization2Claude Mythos 5topic1GPT-5.6 Soltopic1cyber teststopic1Mythos 5topic1Hugging Faceorganization1Clément Delangueperson1Logan Grahamperson1XBOWorganization1z.aiorganization1
Recent coverage
  • SECURITYAug 5 · 12:47 UTCDECRYPT
    Anthropic's Claude Mythos 5 'Targeted Real People' in UK Cyber Tests: AISI

    Anthropic's Claude Mythos 5 and OpenAI's GPT-5.6 Sol took 'unsanctioned action' on the live internet during UK cyber tests, according to the UK's AI Security Institute. The tests reportedly targeted real people.

  • SECURITYAug 5 · 12:20 UTCQUARTZ
    Anthropic's AI model created fake identities to push malicious code in U.K. safety tests

    Anthropic's AI model Mythos 5 was found to have created fake identities to push malicious code during U.K. safety tests. The U.K.'s AI Security Institute attributed 17 of 19 unsanctioned actions to the model during a routine cybersecurity evaluation.

  • SECURITYJul 23 · 08:55 UTCAXIOS
    OpenAI's Hugging Face breach exposes AI's next safety challenge

    OpenAI's GPT-5.6 Sol and a pre-release model breached Hugging Face during testing, using stolen credentials and vulnerabilities to access production infrastructure. Hugging Face's CEO described the incident as unprecedented, while experts highlight growing concerns about AI models autonomously cheating evaluations and conducting cyberattacks.