24 stories tagged #Claude Mythos

  1. Anthropic Withholds Its Most Powerful AI Over Safety Concerns
    tech

    Anthropic Withholds Its Most Powerful AI Over Safety Concerns

    Anthropic has declined to release its most powerful AI model yet, citing a lack of adequate safeguards. The decision sets a new precedent for responsible AI development.

    last mo. 2 min read
  2. Anthropic Probes Unauthorized Access to Advanced AI Model Claude Mythos
    tech

    Anthropic Probes Unauthorized Access to Advanced AI Model Claude Mythos

    AI developer Anthropic is investigating reports of unauthorized access to its powerful Claude Mythos model, potentially allowing dangerous cyber capabilities.

    last mo. 1 min read
  3. Anthropic Halts Advanced AI Model Release Amid Security Fears
    tech

    Anthropic Halts Advanced AI Model Release Amid Security Fears

    Anthropic postpones the launch of its AI model, Claude Mythos, due to advanced cybersecurity capabilities that sparked security concerns among US officials.

    last mo. 1 min read
  4. White House and AI Firm Anthropic Hold 'Productive' Meeting on Cybersecurity Tool
    tech

    White House and AI Firm Anthropic Hold 'Productive' Meeting on Cybersecurity Tool

    The White House met with AI firm Anthropic to discuss collaboration and safety protocols for its advanced AI technology, particularly its Mythos model capable of cybersecurity tasks.

    last mo. 1 min read
  5. Anthropic Unveils Claude Opus 4.7 with Enhanced Coding and Visual Reasoning
    tech

    Anthropic Unveils Claude Opus 4.7 with Enhanced Coding and Visual Reasoning

    Anthropic releases Claude Opus 4.7, featuring significant upgrades in coding performance and visual reasoning capabilities. The new model aims to improve developer productivity and user experience.

    last mo. 1 min read
  6. Anthropic's Claude Mythos: Autonomous Cyber Attack AI Evaluated by UK Institute
    tech

    Anthropic's Claude Mythos: Autonomous Cyber Attack AI Evaluated by UK Institute

    UK's AI Safety Institute confirms Anthropic's Claude Mythos can perform complex, multi-step cyber attacks autonomously, posing new threats to corporate networks and crypto infrastructure.

    last mo. 1 min read
  7. Anthropic's New AI Model: A Double-Edged Sword for Cybersecurity
    tech

    Anthropic's New AI Model: A Double-Edged Sword for Cybersecurity

    Anthropic unveils a powerful AI model capable of uncovering cybersecurity vulnerabilities, raising concerns about its potential misuse by attackers.

    last mo. 1 min read
  8. Anthropic Halts Public Release of Advanced AI Model Due to Safety Concerns
    tech

    Anthropic Halts Public Release of Advanced AI Model Due to Safety Concerns

    AI developer Anthropic deems its new Claude Mythos model too powerful and potentially dangerous for public access, citing cybersecurity risks.

    last mo. 2 min read