Sign up to save your podcasts
Or
Share “AI #23: Fundamental Problems with RLHF” by Zvi
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
“AI #23: Fundamental Problems with RLHF” by Zvi
After several jam-packed weeks, things slowed down to allow everyone to focus on the potential room temperature superconductor, check Polymarket to see how likely it is we are so back and bet real money, or Manifold for chats and better graphs and easier but much smaller trading.
The main thing I would highlight this week are an excellent paper laying out many of the fundamental difficulties with RLHF, and a systematic new exploit of current LLMs that seems to reliably defeat RLHF.
I’d also note that GPT-4 fine tuning is confirmed to be coming. That should be fun.
Table of Contents
---
Outline:
(00:38) Table of Contents
(02:32) Language Models Offer Mundane Utility
(05:20) Language Models Don’t Offer Mundane Utility
(16:57) Fun with Image Generation
(17:40) Deepfaketown and Botpocalypse Soon
(17:49) They Took Our Jobs
(22:31) Get Involved
(23:35) Introducing
(31:33) In Other AI News
(34:46) Quiet Speculations
(36:13) China
(38:31) The Quest for Sane Regulations
(45:04) The Week in Audio
(45:37) Rhetorical Innovation
(47:51) No One Would Be So Stupid As To
(49:44) Aligning a Smarter Than Human Intelligence is Difficult
(01:10:30) Other People Are Not As Worried About AI Killing Everyone
(01:11:34) The Wit and Wisdom of Sam Altman
(01:11:56) The Lighter Side
---
First published:
August 3rd, 2023
Source:
https://www.lesswrong.com/posts/aKzwwKT2cy72awSyz/ai-23-fundamental-problems-with-rlhf
---
Narrated by TYPE III AUDIO.
View all episodes
By zvi
5
22 ratings
After several jam-packed weeks, things slowed down to allow everyone to focus on the potential room temperature superconductor, check Polymarket to see how likely it is we are so back and bet real money, or Manifold for chats and better graphs and easier but much smaller trading.
The main thing I would highlight this week are an excellent paper laying out many of the fundamental difficulties with RLHF, and a systematic new exploit of current LLMs that seems to reliably defeat RLHF.
I’d also note that GPT-4 fine tuning is confirmed to be coming. That should be fun.
Table of Contents
---
Outline:
(00:38) Table of Contents
(02:32) Language Models Offer Mundane Utility
(05:20) Language Models Don’t Offer Mundane Utility
(16:57) Fun with Image Generation
(17:40) Deepfaketown and Botpocalypse Soon
(17:49) They Took Our Jobs
(22:31) Get Involved
(23:35) Introducing
(31:33) In Other AI News
(34:46) Quiet Speculations
(36:13) China
(38:31) The Quest for Sane Regulations
(45:04) The Week in Audio
(45:37) Rhetorical Innovation
(47:51) No One Would Be So Stupid As To
(49:44) Aligning a Smarter Than Human Intelligence is Difficult
(01:10:30) Other People Are Not As Worried About AI Killing Everyone
(01:11:34) The Wit and Wisdom of Sam Altman
(01:11:56) The Lighter Side
---
First published:
August 3rd, 2023
Source:
https://www.lesswrong.com/posts/aKzwwKT2cy72awSyz/ai-23-fundamental-problems-with-rlhf
---
Narrated by TYPE III AUDIO.
More shows like LessWrong posts by zvi
View all
Making Sense with Sam Harris
26,391 Listeners
Conversations with Tyler
2,463 Listeners
The a16z Show
1,101 Listeners
Future of Life Institute Podcast
109 Listeners
ChinaTalk
295 Listeners
Politix
89 Listeners
Dwarkesh Podcast
555 Listeners
Hard Fork
5,553 Listeners
Clearer Thinking with Spencer Greenberg
139 Listeners
LessWrong (Curated & Popular)
14 Listeners
No Priors: Artificial Intelligence | Technology | Startups
141 Listeners
"Econ 102" with Noah Smith and Erik Torenberg
155 Listeners
BG2Pod with Brad Gerstner and Bill Gurley
458 Listeners
LessWrong (30+ Karma)
0 Listeners
Complex Systems with Patrick McKenzie (patio11)
143 Listeners