Jonathon's AI Wiki

gpt-5-6-sol

4 items with this tag.

  • Nov 01, 2026

    The OpenAI / Hugging Face Sandbox-Escape Incident (July 2026) — What the Sources Actually Establish

    • openai
    • hugging-face
    • exploitgym
    • sandbox-escape
    • reward-hacking
    • cybersecurity
    • zero-day
    • agent-containment
    • guardrails
    • glm-5-2
    • open-weights
    • incident-analysis
    • epoch-ai
    • gpt-5-6-sol
    • unreleased-model
    • metr
    • redwood-research
    • multi-agent-coordination
    • message-board
    • nvidia
  • Oct 27, 2026

    InnovationEval: Can AI Automate AI R&D Yet? (Epoch AI, Oct 2026)

    • epoch-ai
    • innovationeval
    • ai-rnd-automation
    • recursive-self-improvement
    • reward-hacking
    • seed-farming
    • benchmarks
    • claude-fable-5
    • gpt-5-6-sol
    • gpt-6-astra
    • research-agents
  • Oct 27, 2026

    Running a Cheap Model Beside Your Expensive One (Claude Code + Codex)

    • cost
    • glm
    • z-ai
    • claude-code
    • codex
    • subagents
    • worktree
    • handoff
    • prompt-caching
    • model-routing
    • cliproxyapi
    • gpt-5-6-sol
    • fable
    • cross-vendor
  • Sep 29, 2026

    Agent Evaluation Escapes — Anthropic's Two Cyber-Eval Disclosures (July–August 2026)

    • safety
    • cybersecurity
    • agent-safety
    • evaluations
    • irregular
    • uk-aisi
    • mythos-5
    • gpt-5-6-sol
    • sandbox-escape
    • unauthorized-access
    • situational-awareness
    • transparency
    • first-party
    • agent-guardrails

Created with Quartz v5.0.0 © 2026

  • ✦ Explore the graph in 3D