🌌

Ilya Sutskever

Original nameIlya Sutskever

Russian-Canadian Computer Scientist, Co-Founder of OpenAI

The Math Behind the Algorithm · Theoretical Founders

AlexNetThe "Scaling Law" Idea of the GPT SeriesOpenAI

Who they are

Ilya Sutskever (1986– ) is a computer scientist born in Russia and raised in Israel and Canada, who studied under Geoffrey Hinton during his doctorate at the University of Toronto and, with Alex Krizhevsky, designed AlexNet, the 2012 ImageNet competition winner, a result that directly triggered the wave of the deep-learning revival. Sutskever in 2015 took part as a co-founder in founding OpenAI, long serving as chief scientist, one of the most important early advocates of the "scaling law" idea — that as long as one keeps increasing model parameter scale, training-data volume, and compute, a language model’s capability will predictably rise — an idea that thereafter became the core guiding thought of the R&D strategy of the GPT series. During the November 2023 OpenAI board crisis, Sutskever as a board member voted to support removing Sam Altman as CEO, a stance that thereafter provoked wide controversy, and five days later, with Altman’s reinstatement, Sutskever publicly wrote that he regretted taking part in it. In May 2024 he announced his departure from OpenAI and then founded the company "Safe Superintelligence," focused on AI-safety research.

Primary sourcesKrizhevsky, Sutskever & Hinton, "ImageNet Classification with Deep Convolutional Neural Networks" (2012)

Key stories

A System That Changed the Whole Field’s Mind

The AlexNet Sutskever and Krizhevsky designed under their adviser Hinton won the 2012 ImageNet competition by an enormous margin over the runner-up, its image-recognition error rate falling sharply below the previous best, and this overwhelming victory made the scholarly researchers who had widely doubted deep learning’s practical value turn almost unanimously to this technical path within a very short time, and Sutskever himself rose swiftly from a doctoral student to one of the most watched young scholars of the field.

ChatGPT’s Hundred Million Users in Two Months, a CEO Removed and Reinstated in Five Days

OpenAI, which Sutskever co-founded and long served as chief scientist, released ChatGPT in November 2022, its users breaking a hundred million in two months; in November 2023 he and other board members for a time suddenly removed Altman as CEO, and five days later, under a joint employee protest, Altman was reinstated, and Sutskever publicly wrote that he deeply regretted taking part in the vote to remove him. This governance crisis and his later choice to leave OpenAI and found Safe Superintelligence are a classic case for examining the internal line divergence of this wave of the AI industry.

Relationships

Mentor

🧠Geoffrey Hinton

Ally

🧯Jan Leike

Nemesis

💬Sam Altman

Echoes today

Below are how modern works borrow or reinterpret this name or story — not the original material. The two differ, so keep them apart.

From "scaling law" believer to founder of "Safe Superintelligence"Sutskever’s path from OpenAI chief scientist to leaving to found a company focused on safety research was thereafter widely seen as an important individual case for examining the internal philosophical divergence of the AI industry.

Appears in

The Deep-Learning Revival and the Shock of ImageNetPersistence Through the "Neural-Network Winter" and the Shock of AlexNet The Generative-AI Explosion and the Industry ContestChatGPT Past a Hundred Million Users in Two Months, a CEO Fired and Reinstated in Five Days

← Back to Figures