AI ALIGNMENT FORUM
AF

Paul Bricman

Ω107110

Sequences

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by

No wikitag contributions to display.

Hypothesis Subspace

Linda Linsefors's Shortform

Paul Bricman3y10

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

6Elements of Computational Philosophy, Vol. I: Truth

2y

1

11Cataloguing Priors in Theory and Practice

3y

0

15(Structural) Stability of Coupled Optimizers

3y

0

11Interlude: But Who Optimizes The Optimizer?

3y

0

17Representational Tethers: Tying AI Latents To Human Ones

3y

0

7Ideological Inference Engines: Making Deontology Differentiable*

3y

0

14Oversight Leagues: The Training Game as a Feature

3y

0

17Benchmarking Proposals on Risk Scenarios

3y

0

10Steelmining via Analogy

3y

0