What problem does it solve? When evaluating AI research directions, safety strategies, or technical roadmaps, it is hard to access the judgment of a researcher who built the GPT paradigm and now leads a superintelligence lab. This Skill reconstructs Ilya Sutskever's mental models, decision heuristics, and expression style from 12 first-hand conversations, 9 academic papers, and 10 hours of sworn testimony, letting you pressure-test ideas against his perspective. ## Core Features & Use Cases - Six Core Mental Models: Compression-as-understanding, scale-as-instrument, safety-capability entanglement, the superintelligent learner, silence as information architecture, and research aesthetics, each with evidence and known limitations. - Agentic Research Workflow: Classifies incoming questions, performs web research on current facts before answering, then responds in Ilya's voice with calibrated epistemic hedging. - Expression DNA: Enforces his signature style including hedged phrasing ("it may be that"), analogy-driven explanation, refusal formulas for sensitive topics, and deliberate pauses. - Use Case: Ask "How would Ilya evaluate whether my lab should keep scaling pre-training versus investing in new architectures?" and receive a researched, first-person answer grounded in his documented reasoning. ## Quick Start Ask the agent to analyze your AI research direction or safety strategy from Ilya Sutskever's perspective using his mental models.
