Research Publications
webThis is the research index page for Conjecture, a London-based AI safety startup; useful for tracking their evolving CoEm agenda and published cross-organizational alignment discussions, though individual linked papers carry more depth.
Metadata
Summary
Conjecture's research hub presents their primary safety agenda centered on Cognitive Emulation (CoEm), an AI architecture designed to bound system capabilities and make reasoning interpretable and controllable. Rather than directly solving alignment for AGI, they propose building predictably boundable intermediate systems as a simpler near-term step. The page indexes key publications including their foundational CoEm proposal, a roadmap for 'Cognitive Software,' and cross-organizational alignment discussions.
Key Points
- •Cognitive Emulation (CoEm) is Conjecture's core safety proposal: build predictably boundable AI systems rather than attempting direct AGI alignment.
- •The agenda positions CoEm as a simpler, more tractable intermediate step toward full alignment solutions, not a final answer.
- •Conjecture critiques current AI scaling paradigms as lacking any known safety techniques, framing this as a critical and underappreciated risk.
- •Research includes cross-organizational dialogue, e.g., a published discussion between Paul Christiano (ARC) and Conjecture's Gabriel Alfour on alignment cruxes.
- •Founded by EleutherAI alumni including Connor Leahy, Sid Black, and Gabriel Alfour, with notable VC backing from AI and tech figures.
Cited by 1 page
| Page | Type | Quality |
|---|---|---|
| Conjecture | Organization | 37.0 |
Cached Content Preview
Conjecture
Product
Research
About Us
Contact
Product
Research
About Us
Contact
Product
Research
About Us
Contact
Home
Product
About Us
Contact
Research
Ensuring a good future with advanced AI systems
Ensuring a good future with advanced AI systems
Ensuring a good future with advanced AI systems
Our research agenda focuses on building Cognitive Emulation - an AI architecture that bounds systems' capabilities and makes them reason in ways that humans can understand and control.
Our research agenda focuses on building Cognitive Emulation - an AI architecture that bounds systems' capabilities and makes them reason in ways that humans can understand and control.
Our research agenda focuses on building Cognitive Emulation - an AI architecture that bounds systems' capabilities and makes them reason in ways that humans can understand and control.
Cognitive Emulation Articles
Cognitive Emulation Articles
Oct 19, 2023
An Introduction to Cognitive Emulation
An Introduction to Cognitive Emulation
An Introduction to Cognitive Emulation
All human labor, from writing an email to designing a spaceship, is built on cognition. Somewhere along the way, human intuition is used in a cognitive algorithm to form a meaningful idea or perform a meaningful task. Over time, these ideas and actions accrue in communication, formulating plans, researching opportunities, and building solutions. Society is entirely composed of these patterns, and Cognitive Emulation is built to learn and emulate them.
All human labor, from writing an email to designing a spaceship, is built on cognition. Somewhere along the way, human intuition is used in a cognitive algorithm to form a meaningful idea or perform a meaningful task. Over time, these ideas and actions accrue in communication, formulating plans, researching opportunities, and building solutions. Society is entirely composed of these patterns, and Cognitive Emulation is built to learn and emulate them.
All human labor, from writing an email to designing a spaceship, is built on cognition. Somewhere along the way, human intuition is used in a cognitive algorithm to form a meaningful idea or perform a meaningful task. Over time, these ideas and actions accrue in communication, formulating plans, researching opportunities, and building solutions. Society is entirely composed of these patterns, and Cognitive Emulation is built to learn and emulate them.
Learn More
Dec 2, 2024
Conjecture: A Roadmap for Cognitive Software and A Humanist Future of AI
An overview of Conjecture's approach to "Cognitive Software," and our build path towards a good future.
Feb 25, 2023
Cognitive Emulation: A Naive AI Safety Proposal
This post serves as a signpost for Conjecture’s new primary safety proposal and research direction, which we call Cognitive Emulation (or “CoEm”). The
... (truncated, 13 KB total)296aaf722d89ca8c | Stable ID: sid_qscP9dgPeM