Open Agency Architecture

TagLast edit: Apr 10, 2024, 10:53 PM by Chipmonk

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.

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

Charbel-Raphaël and Gabin

Apr 19, 2023, 4:09 PM

168 points

40 comments21 min readLW link 2 reviews

An Open Agency Architecture for Safe Transformative AI

davidadDec 20, 2022, 1:04 PM

80 points

22 comments4 min readLW link

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

davidadAug 26, 2023, 3:12 PM

165 points

29 comments5 min readLW link

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

Roman Leventov and Rafael Kaufmann Nedal

Dec 20, 2023, 5:11 PM

22 points

8 comments16 min readLW link

The Open Agency Model

Eric DrexlerFeb 22, 2023, 10:35 AM

114 points

18 comments4 min readLW link

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

Roman LeventovDec 19, 2023, 4:49 PM

17 points

5 comments3 min readLW link

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

Roman LeventovDec 27, 2023, 2:51 PM

33 points

9 comments4 min readLW link

Provably Safe AI: Worldview and Projects

Ben Goldhaber and Steve_Omohundro

Aug 9, 2024, 11:21 PM

54 points

44 comments7 min readLW link

Roadmap for a collaborative prototype of an Open Agency Architecture

Deger TuranMay 10, 2023, 5:41 PM

31 points

0 comments12 min readLW link

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

ChipmonkJan 3, 2024, 5:55 PM

48 points

3 comments3 min readLW link

Apply to the Conceptual Boundaries Workshop for AI Safety

ChipmonkNov 27, 2023, 9:04 PM

50 points

0 comments3 min readLW link

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

Joar SkalseMay 17, 2024, 7:13 PM

67 points

10 comments2 min readLW link

«Boundaries/Membranes» and AI safety compilation

ChipmonkMay 3, 2023, 9:41 PM

56 points

17 comments8 min readLW link

Davidad’s Provably Safe AI Architecture—ARIA’s Programme Thesis

simeon_cFeb 1, 2024, 9:30 PM

69 points

17 comments1 min readLW link

(www.aria.org.uk)

What does davidad want from «boundaries»?

Chipmonk and davidad

Feb 6, 2024, 5:45 PM

47 points

1 comment5 min readLW link

Announcing Atlas Computing

miyazonoApr 11, 2024, 3:56 PM

44 points

4 comments4 min readLW link

Why I find Davidad’s plan interesting

Paul WMay 20, 2024, 8:13 AM

18 points

0 comments6 min readLW link

No comments.