AI ALIGNMENT FORUM
AF

Paul Bricman

Sequences

Hypothesis Subspace

Posts

Sorted by New

6Elements of Computational Philosophy, Vol. I: Truth

1y

1

11Cataloguing Priors in Theory and Practice

2y

0

15(Structural) Stability of Coupled Optimizers

2y

0

11Interlude: But Who Optimizes The Optimizer?

2y

0

17Representational Tethers: Tying AI Latents To Human Ones

2y

0

7Ideological Inference Engines: Making Deontology Differentiable*

2y

0

14Oversight Leagues: The Training Game as a Feature

2y

0

17Benchmarking Proposals on Risk Scenarios

2y

0

10Steelmining via Analogy

2y

0

Wiki Contributions

Comments

Sorted by

Linda Linsefors's Shortform

Paul Bricman2y10

You mean, in that you can simply prompt for a reasonable non-infinite performance and get said outcome?