LESSWRONGTags
LW

Refine

EditHistorySubscribe

Help improve this page

EditHistorySubscribe

Help improve this page

Contributors

Refine is a conceptual research incubator hosted by Conjecture.

Posts tagged Refine

4

39confusion about alignment requirements

1y

10

4

19Refine's Second Blog Post Day

1y

0

3

32What if we approach AI safety like a technical engineering safety problem

1y

4

3

25Benchmarking Proposals on Risk Scenarios

1y

2

2

107What I Learned Running Refine

10mo

5

2

50PreDCA: vanessa kosoy's alignment protocol

1y

8

2

39the Insulated Goal-Program idea

1y

4

2

30Representational Tethers: Tying AI Latents To Human Ones

1y

0

2

30Epistemic Artefacts of (conceptual) AI alignment research

Nora_Ammann, particlemania

1y

1

2

27goal-program bricks

1y

2

2

27ordering capability thresholds

1y

0

2

23Refine Blogpost Day #3: The shortforms I did write

Alexander Gietelink Oldenziel

1y

0

2

18Refine's Third Blog Post Day/Week

1y

0

1

143Refine: An Incubator for Conceptual Alignment Research Bets

1y

13

1

87How to Diversify Conceptual Alignment: the Model Behind Refine

1y

11