Open Agency Architecture

Edited by Chris Lakin, Roman Leventov, niplav last updated 10th Apr 2024

The Open Agency Architecture ("OAA") is an AI alignment proposal by (among others) @davidad and @Eric Drexler.

See Davidad's Provably Safe AI Architecture - ARIA's Programme Thesis for the most up-to-date (2024 Feb 1) and quite readable explanation.

A shorter but older explanation is also available here.

New better link: https://www.aria.org.uk/programme-safeguarded-ai/

Atlas Computing is the org intended to house OAA.
Gaia Network is a variant of an open agency architecture. Gaia Network is related to (the) OAA, but not directly descending from it.

Posts tagged Open Agency Architecture

6

169Davidad's Bold Plan for Alignment: An In-Depth Explanation

Ω

Charbel-Raphaël, Gabin

3y

Ω

40

5

165A list of core AI safety problems and how I hope to solve them

Ω

davidad

2y

Ω

29

5

80An Open Agency Architecture for Safe Transformative AI

Ω

davidad

3y

Ω

22

3

114The Open Agency Model

Ω

Eric Drexler

3y

Ω

19

3

22Gaia Network: a practical, incremental pathway to Open Agency Architecture

Roman Leventov, Rafael Kaufmann Nedal

2y

8

2

33AGI will be made of heterogeneous components, Transformer and Selective SSM blocks will be among them

Ω

Roman Leventov

2y

Ω

9

2

17SociaLLM: proposal for a language model design for personalised apps, social science, and AI safety research

Roman Leventov

2y

5

1

69Davidad's Provably Safe AI Architecture - ARIA's Programme Thesis

Ω

simeon_c

2y

Ω

17

1

67Towards Guaranteed Safe AI: A Framework for Ensuring Robust and Reliable AI Systems

Ω

Joar Skalse

2y

Ω

10

1

56«Boundaries/Membranes» and AI safety compilation

Chris Lakin

3y

17

1

54Provably Safe AI: Worldview and Projects

Ben Goldhaber, Steve_Omohundro

1y

44

1

50Apply to the Conceptual Boundaries Workshop for AI Safety

Chris Lakin

2y

0

1

48Safety First: safety before full alignment. The deontic sufficiency hypothesis.

Ω

Chris Lakin

2y

Ω

3

1

47What does davidad want from «boundaries»?

Ω

Chris Lakin, davidad

2y

Ω

1

45Announcing Atlas Computing

miyazono

2y

4

LESSWRONG
LW

LESSWRONG
LW

Open Agency Architecture