Connor Leahy

Connor Leahy est un chercheur en apprentissage automatique qui a cofondé EleutherAI et dirigé le laboratoire de sécurité Conjecture jusqu’en 2026 ; il dirige aujourd’hui ControlAI aux États-Unis et soutient que la construction d’une superintelligence devrait être interdite comme risque d’extinction.

Fonctions

juil. 2020 – aujourd’huiCo-founder and co-lead, EleutherAIGrassroots collective that released The Pile, GPT-Neo, GPT-J and GPT-NeoX.Source →
mars 2022 – mars 2026Founder and Chief Executive Officer, ConjectureLondon AI-safety lab; wound down in March 2026 because frontier research had become too expensive.Source →
mars 2026 – aujourd’huiUS Executive Director, ControlAIBased in Washington DC, briefing lawmakers on extinction risk from superintelligence.Source →

Au registre

20 août 2026 · Entretien · Democracy Now!
We are developing systems that are not tools. They are not weapons. They are adversaries. … we should criminalize the creation of superintelligence, the same way that it is illegal to build a nuclear bomb.
US Executive Director, ControlAI à l’époque
AI Superintelligence Is Not a Tool, It's an Adversary Threatening Humanity: ControlAI's Connor Leahy →
26 mars 2026 · Publication · Expedition to the Far Lands (Substack)
The problem with ASI is not that ASI is killing us. ASI doesn't exist, yet. It's that we are allowing unaccountable corporations to build things that threaten the lives of us and everyone we love.
US Executive Director, ControlAI à l’époque
Written on closing Conjecture and moving to ControlAI.
Conjecture: A Retrospective →
1 déc. 2025 · Audition · House of Commons of Canada, Standing Committee on Access to Information, Privacy and Ethics
Recently, AI companies have been racing to automate AI research itself, allowing AIs to build even better AIs by themselves. This process is called recursive self-improvement, meaning the moment an AI is built that is good enough to make better AIs, it might already be too late.
Founder and Chief Executive Officer, Conjecture à l’époque
Appeared as CEO of Conjecture Ltd.
Evidence - ETHI (45-1) - No. 20 →
31 oct. 2024 · Article scientifique · The Compendium
Humanity risks extinction from its very creations — AIs. … If they succeed in creating AIs outsmarting humanity, it would be game over -- no one knows how to keep control of these AIs.
Founder and Chief Executive Officer, Conjecture à l’époque
Co-written with Gabriel Alfour, Chris Scammell, Andrea Miotti and Adam Shimi; a living document first published 31 October 2024.
The Compendium →
17 févr. 2023 · Entretien · TIME
These systems, as they become more powerful, are not becoming less alien. If anything, we're putting a nice little mask on them with a smiley face.
Founder and Chief Executive Officer, Conjecture à l’époque
Speaking as CEO of Conjecture about the Bing chatbot; he added "We cannot bound what these systems do at all."
The New AI-Powered Bing Is Threatening Users. That's No Laughing Matter →