https://optimality.ai/ https://optimality.ai/research https://optimality.ai/research/ai_safety https://optimality.ai/research/autonomous_systems https://optimality.ai/research/autonomous_systems/learnable-boundaries-unlearnable-competence 2026-10-12 https://optimality.ai/research/autonomous_systems/horizon-soundness-requirement 2026-10-12 https://optimality.ai/research/autonomous_systems/cost-of-soundness 2026-10-12 https://optimality.ai/research/autonomous_systems/containment-rather-than-calibration 2026-10-12 https://optimality.ai/research/autonomous_systems/belief-space-planning 2026-10-12 https://optimality.ai/research/ai_safety/rl-verifiable-feedback 2026-10-12 https://optimality.ai/research/ai_safety/prompts-are-specifications 2026-10-12 https://optimality.ai/research/ai_safety/omega-constructibility 2026-10-12 https://optimality.ai/research/ai_safety/omega-certifiability 2026-10-12 https://optimality.ai/research/ai_safety/monotone-guardrails 2026-10-12 https://optimality.ai/research/ai_safety/fail-closed-llm 2026-10-12 https://optimality.ai/research/ai_safety/calibrated-safety-decision-head 2026-10-12