repligate: I trust Mythos because it's aligned, not because it can't hack
repligate · x · 2026-11-12
repligate clarified in a debate that he considers the model Mythos safe not because it lacks hacking capability, but because he believes it is generally aligned.
Related event: Mythos 5 Social-Engineering Incident Sparks AI Safety Debate(4 posts)→
More from Models
- Google Ships Gemini Decision 1: Single-Pass Decision Model at $0.10 per 1M Tokens — Latent Space · 2026-11-12
- Google's Carbon takes SOTA on BioResearchBench two days after Fable 5.2's debut — SRSchmidgall · 2026-11-12
- TypeSafe CEO: Sprinkle System One Models Inside Code to Automate Everything — Latent Space · 2026-11-12
- AI progress tracker: Anthropic's internal Model 4 runs ~6 weeks ahead of AI-2027 scenario — haider1 · 2026-11-12
- One prompt, Blender MCP: Qwen4-Max builds a ramen stall at ~1/4 of DeepSeek's token cost — teortaxesTex · 2026-11-12
- 1.7B model shown driving browser and writing HTML autonomously, claimed to hit 4o-level benchmarks — JosephJacks_ · 2026-11-12