The Bayesian Conspiracy

179 – The Plan (to align AI), with John Wentworth

Jan 25, 2023

01:21:10

Creator website

Highlights

AI Chapters

Episode notes

The Quarks Simulation on Apple Plus

02:28

Is Iterative Engineering a Good Idea?

02:08

How's the Thermometer of Abstraction Coming?

03:08

Introduction

5min

Is E. Coli Agent T?

3min

Is JAT GPT the Hot New Thing?

3min

Is There a Smoke Alarm for AI?

2min

Is Iterative Engineering a Good Idea?

2min

Neural Networks - What Are We Fundamentally Confusioned About?

3min

E. Coli Models Are Different Than Your Thermostat Models

5min

Is There Significantly More People Working on These Problems?

2min

Is Stable Diffusion Really a Ball?

4min

10.

Why Do We Need Formalizations for Engineering?

4min

11.

Is There a Phase Transition?

4min

12.

The City Wouldn't Give It a 50% Chance

5min

13.

How's the Thermometer of Abstraction Coming?

3min

14.

Is There a Doom Mindset?

2min

15.

What's the Basic Foundation Sequence?

5min

16.

Is There Anything We Can Do About MATS Training?

3min

17.

The Deal to the Rose Is Doing

2min

18.

You Can Set Goals Anytime, Do It Now

2min

19.

Eliezer's Unweaving the Rainbow

3min

20.

The Emergent Beauty in Math

2min

21.

Is Reality Uglier Than Mechanics?

2min

22.

How to Cool a Scoop of Hot Soup

2min

23.

How Do I Translate the Quarks?

2min

24.

The Quarks Simulation on Apple Plus

2min

25.

The Uncertainty of Gene Expression Networks

2min

26.

Is Reality Ugly in Quotes?

2min

27.

Is Reality Ugly?

2min

28.

Does That Mean That Reality Is Overall Ugly?

2min

29.

The Mind Killer Podcast

2min

Worried about AGI running amok in the near future? John Wentworth thinks we can align AI. In 10-15 years. With greater than 50/50 probability. And he has a plan! We discuss The Plan, its merits, and how it has progressed … Continue reading →

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.