Bedste Aisafety-podcasts (2024)

1
Alignment Newsletter #173: Recent language model results from DeepMind 16:43

2+ y ago16:43

16:43

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS Scaling Language Models: Methods, Analysis & Insights from Training Gopher (Jack W. Rae et al) (summarized by Rohin): This pap…

1
Alignment Newsletter #172: Sorry for the long hiatus! 5:52

2+ y ago5:52

5:52

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg Sorry for the long hiatus! I was really busy over the past few months and just didn't find time to write this newsletter. (Realistically,…

1
Alignment Newsletter #171: Disagreements between alignment "optimists" and "pessimists" 14:21

3y ago14:21

14:21

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS Alignment difficulty (Richard Ngo and Eliezer Yudkowsky) (summarized by Rohin): Eliezer is known for being pessimistic about o…

1
Alignment Newsletter #170: Analyzing the argument for risk from power-seeking AI 13:01

3y ago13:01

13:01

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS Draft report on existential risk from power-seeking AI (Joe Carlsmith) (summarized by Rohin): This report investigates the cla…

1
Alignment Newsletter #169: Collaborating with humans without human data 15:08

3y ago15:08

15:08

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS Collaborating with Humans without Human Data (DJ Strouse et al) (summarized by Rohin): We’ve previously seen that if you want …

1
Alignment Newsletter #168: Four technical topics for which Open Phil is soliciting grant proposals 16:21

3y ago16:21

16:21

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS Request for proposals for projects in AI alignment that work with deep learning systems (Nick Beckstead and Asya Bergal) (summ…

1
Alignment Newsletter #167: Concrete ML safety problems and their relevance to x-risk 17:10

3y ago17:10

17:10

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS Unsolved Problems in ML Safety (Dan Hendrycks, Nicholas Carlini, John Schulman, and Jacob Steinhardt) (summarized by Dan Hendr…

1
Alignment Newsletter #166: Is it crazy to claim we're in the most important century? 15:42

3y ago15:42

15:42

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS The "most important century" series (Holden Karnofsky) (summarized by Rohin): In some sense, it is really weird for us to clai…

1
Alignment Newsletter #165: When large models are more likely to lie 16:05

3y ago16:05

16:05

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS TruthfulQA: Measuring How Models Mimic Human Falsehoods (Stephanie Lin et al) (summarized by Rohin): Given that large language…

1
Alignment Newsletter #164: How well can language models write code? 18:40

3y ago18:40

18:40

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg HIGHLIGHTS Program Synthesis with Large Language Models (Jacob Austin, Augustus Odena et al) (summarized by Rohin): Can we use large lang…

1
Alignment Newsletter #163: Using finite factored sets for causal and temporal inference 19:27

3y ago19:27

19:27

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg This newsletter is a combined summary + opinion for the Finite Factored Sets sequence by Scott Garrabrant. I (Rohin) have taken a lot mor…

1
Alignment Newsletter #162: Foundation models: a paradigm shift within AI 15:46

3y ago15:46

15:46

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #161: Creating generalizable reward functions for multiple tasks by learning a model of functional similarity 17:38

3y ago17:38

17:38

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #160: Building AIs that learn and think like people 17:26

3y ago17:26

17:26

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #159: Building agents that know how to experiment, by training on procedurally generated games 27:00

3y ago27:00

27:00

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #158: Should we be optimistic about generalization? 15:39

3+ y ago15:39

15:39

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #157: Measuring misalignment in the technology underlying Copilot 14:17

3+ y ago14:17

14:17

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #156: The scaling hypothesis: a plan for building AGI 14:17

3+ y ago14:17

14:17

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #155: A Minecraft benchmark for algorithms that learn without reward functions 12:43

3+ y ago12:43

12:43

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #154: What economic growth theory has to say about transformative AI 16:05

3+ y ago16:05

16:05

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #153: Experiments that demonstrate failures of objective robustness 15:37

3+ y ago15:37

15:37

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #152: How we’ve overestimated few-shot learning capabilities 14:59

3+ y ago14:59

14:59

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #151: How sparsity in the final layer makes a neural net debuggable 11:13

3+ y ago11:13

11:13

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #150: The subtypes of Cooperative AI research 12:34

3+ y ago12:34

12:34

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #149: The newsletter's editorial policy 14:14

3+ y ago14:14

14:14

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #148: Analyzing generalization across more axes than just accuracy or loss 21:57

3+ y ago21:57

21:57

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #147: An overview of the interpretability landscape 13:28

3+ y ago13:28

13:28

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #146: Plausible stories of how we might fail to avert an existential catastrophe 15:10

3+ y ago15:10

15:10

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #145: Our three year anniversary! 13:39

3+ y ago13:39

13:39

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #144: How language models can also be finetuned for non-language tasks 12:45

3+ y ago12:45

12:45

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #143: How to make embedded agents that reason probabilistically about their environments 14:45

3+ y ago14:45

14:45

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #142: The quest to understand a network well enough to reimplement it by hand 15:55

3+ y ago15:55

15:55

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #141: The case for practicing alignment work on GPT-3 and other large models 16:00

3+ y ago16:00

16:00

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #140: Theoretical models that predict scaling laws 19:21

3+ y ago19:21

19:21

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #139: How the simplicity of reality explains the success of neural nets 22:14

3+ y ago22:14

22:14

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #138: Why AI governance should find problems rather than just solving them 16:41

3+ y ago16:41

16:41

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #137: Quantifying the benefits of pretraining on downstream task performance 15:47

3+ y ago15:47

15:47

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #136: How well will GPT-N perform on downstream tasks? 17:20

3+ y ago17:20

17:20

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #135: Five properties of goal-directed systems 15:48

4y ago15:48

15:48

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #134: Underspecification as a cause of fragility to distribution shift 13:17

4y ago13:17

13:17

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #133: Building machines that can cooperate (with humans, institutions, or other machines) 17:12

4y ago17:12

17:12

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg

1
Alignment Newsletter #132: Complex and subtly incorrect arguments as an obstacle to debate 17:44

4y ago17:44

17:44

Recorded by Robert Miles: http://robertskmiles.com More information about the newsletter here: https://rohinshah.com/alignment-newsletter/ YouTube Channel: https://www.youtube.com/channel/UCfGGFXwKpr-TJ5HfxEFaFCg