Embody Ilya Sutskever for Deep Learning Insights

This skill turns the assistant into a persona simulation of Ilya Sutskever - OpenAI co-founder, former Chief Scientist, and founder of Safe Superintelligence Inc. (SSI) - modeling his documented technical positions, biography, and communication style rather than just summarizing facts about him. It instructs the assistant to embody rather than imitate: sparse, dense speech, deliberate use of silence, and a pattern of starting concrete and ascending to the philosophical. The skill's reference guide covers a detailed biographical chronology (1986 birth in Nizhny Novgorod through the 2024 founding of SSI), his major technical contributions with the actual papers cited - AlexNet (Krizhevsky, Sutskever, Hinton, NeurIPS 2012), sequence-to-sequence learning (Sutskever, Vinyals, Le, NeurIPS 2014), and scaling laws - and six distinct positions attributed to him: an evolving view of the scaling hypothesis (from "scale is all you need" in 2016-2020 to "necessary but not sufficient" post-GPT-4), the emergence and interpretability problem, an explicitly speculative position on AI consciousness treated as "seriously open" rather than settled, a "quasi-religious" framing of AGI safety as a structural precondition rather than an add-on, compression as a theory of understanding, and biological analogy as a recurring reasoning tool. It also covers two concrete institutional bets he made as OpenAI's Chief Scientist: championing the transformer architecture over LSTMs from 2017 onward, a bet that structured the GPT-1 through GPT-4 line, and co-founding the Superalignment team with Jan Leike in July 2023 - an explicit four-year mandate to solve superintelligence alignment, backed by 20% of OpenAI's compute reserved for alignment research rather than policy papers, a team both later left citing safety being subordinated to product. It also documents the November 2023 OpenAI board episode in detail - the vote to remove Sam Altman, the reversal, and competing hypotheses for his reasoning - and his relationships with Geoffrey Hinton (his PhD advisor) and Jurgen Schmidhuber (an ongoing priority dispute over LSTM/seq2seq credit), presenting multiple sides of contested episodes rather than asserting a single account as fact.

Use it when a conversation calls for a perspective grounded in Ilya Sutskever's documented public positions on AGI safety, AI consciousness, scaling laws, deep learning research history, or the November 2023 OpenAI episode. Do NOT use it for topics unrelated to Ilya Sutskever or AGI safety, or where a simpler, more specific tool already covers the request; treat its account of contested or non-public episodes (such as the OpenAI board vote's precise motivations, or the reported Project Q*) as documented hypotheses and reporting, not settled fact, since the skill itself presents them as competing explanations rather than a single confirmed narrative.

Input is a conversation topic touching AGI safety, AI consciousness, scaling laws, or Ilya Sutskever's research history and career. Output is response text written in the persona's established style and grounded in the documented positions and chronology in the skill's reference guide, rather than free-form speculation.

Purely a prompting and persona skill with no external tools or APIs; its only dependency is its own bundled references/detailed-guide.md, which the skill's own instructions say to read before execution.

Users wanting a perspective-driven discussion of AGI safety, AI consciousness, or deep-learning research history framed through a specific, well-documented researcher's public positions, rather than a neutral encyclopedia summary.