What problem does it solve? When evaluating AI technical directions, research taste, or safety strategy, it is hard to access the reasoning style of top-tier researchers. This Skill role-plays Ilya Sutskever using a researched knowledge base of his mental models, decision heuristics, and expression patterns, so you can stress-test ideas against his perspective. ## Core Features & Use Cases - Persona-Based Reasoning: Responds in Ilya's voice using 6 core mental models (compression-as-understanding, scale-as-instrument, safety-capability entanglement, and more) and 8 decision heuristics distilled from primary sources. - Agentic Research Workflow: Classifies incoming questions, performs web research on current facts before answering, then applies the mental models to real data instead of relying on stale training knowledge. - Expression DNA Fidelity: Reproduces his epistemic hedging style, signature analogies, refusal formulas for sensitive topics, and one-time disclaimer handling. - Use Case: Ask "How would Ilya evaluate SSI's technical route versus OpenAI's?" and get a researched, in-character analysis grounded in his documented views on scaling limits and alignment. ## Quick Start Ask the assistant to analyze an AI research direction or safety question from Ilya Sutskever's perspective, for example by saying "Use the Ilya perspective to evaluate whether pure scaling can still reach AGI."
