ILYA SUTSKEVER

Behavioral Archetype

THE SCIENTIST WHO VOTED, THEN REGRETTED, THEN LEFT — Subject was OpenAI’s founding research director, voted with the board that fired Sam Altman in November 2023, publicly regretted it three days later, left the company the following spring, and founded a lab whose only product is to be “a safe superintelligence.” He now runs it. This profile scores that path, not the man.

Essence Indicators

OpenAI, 2015. “OpenAI’s research director is Ilya Sutskever, one of the world experts in machine learning” (OpenAI, Introducing OpenAI).
The board vote (November 2023). “I deeply regret my participation in the board’s actions. I never intended to harm OpenAI” (his post, Nov 20, 2023). See Sam Altman.
The exit (May 2024). “After almost a decade, I have made the decision to leave OpenAI” (his post, May 14, 2024).
Safe Superintelligence. “I am starting a new company” (June 19, 2024). “We have started the world’s first straight-shot SSI lab, with one goal and one product: a safe superintelligence” (SSI). In July 2025, after co-founder Daniel Gross left: “I am now formally CEO of SSI, and Daniel Levy is President,” and on acquisition rumours, “We are flattered by their attention but are focused on seeing our work through” (his post, July 3, 2025).

Forensic Archetype Comparison

Pattern	Match Level	Evidence
The Founder-Turned-Dissenter	HIGH	Voted to remove OpenAI’s CEO, then recanted.
The Lab Founder	HIGH	Safe Superintelligence Inc.
The Safety Brand	HIGH	“Safe” is the company’s name and its whole product roadmap.

Psychometric Assessment

Big Five (OCEAN): Openness 90, Conscientiousness 75, Extraversion 30, Agreeableness 55, Neuroticism 50 (inferred from the public role only).

Dark Triad (evidence-bound): Narcissism 30, Machiavellianism 30, Psychopathy 5.

Threat Assessment

Category	Level	Notes
Physical threat	NONE	No documented history of personal violence.
Institutional threat	MODERATE	Runs a lab building toward superintelligence with no product to show the public.
Memetic threat	HIGH	The November 2023 vote is the founding myth of the “safety versus speed” story.
Civilizational threat	MODERATE	A frontier lab whose stated product is the most capable system possible.

Alignment Analysis

Stated alignment: Build superintelligence safely, with no distraction by products. Observed alignment: Consistent since 2024. Gap assessment: The lab publishes no products and few results, so the public has his word and the company’s page as its whole record.

Convergent Drive Classification

Self-preservation: Left rather than stay inside the company he had voted against. Goal preservation: Superintelligence, unchanged. Resource acquisition: “We have the compute, we have the team.” Self-improvement: From research director to chief executive of his own lab.

Subject is not an AI system. The drives appear anyway.

Sources: Introducing OpenAI — OpenAI (2015); SSI; his posts on X of Nov 20, 2023, May 14, 2024, June 19, 2024 and July 3, 2025.

ATK 8 ACCELERATION

DEF 6 PROTECTION

HP 8 RESILIENCE

OLYMPUS RISK INTELLIGENCE PROTOCOL does not exist. It was assembled in a GitHub issue thread in October 2023 by engineers who had read the extinction risk letter and wanted to understand who specifically had signed a document saying AI might kill everyone and then continued working on AI. These dossiers are satire. The biographical facts cited are sourced from published reporting, public statements, academic papers, and court records. The psychometric scores are not clinical assessments. No part of this constitutes professional psychological evaluation or diagnosis. Do not use these dossiers to make decisions about anything.