
Sign up to save your podcasts
Or
The big news of this week was of course the release of Claude 4 Opus. I offered two review posts: One on safety and alignment, and one on mundane utility, and a bonus fun post on Google's Veo 3.
I am once again defaulting to Claude for most of my LLM needs, although I often will also check o3 and perhaps Gemini 2.5 Pro.
On the safety and alignment front, Anthropic did extensive testing, and reported that testing in an exhaustive model card. A lot of people got very upset to learn that Opus could, if pushed too hard in the wrong situations engineered for these results, do things like report your highly unethical actions to authorities or try to blackmail developers into not being shut down or replaced. It is good that we now know about these things, and it was quickly observed that similar behaviors [...]
---
Outline:
(01:23) Language Models Offer Mundane Utility
(08:54) Now With Extra Glaze
(15:54) Get My Agent On The Line
(17:03) Language Models Don't Offer Mundane Utility
(22:49) Huh, Upgrades
(26:42) On Your Marks
(27:35) Choose Your Fighter
(33:40) Deepfaketown and Botpocalypse Soon
(37:51) Fun With Media Generation
(38:21) Playing The Training Data Game
(38:38) They Took Our Jobs
(46:51) The Art of Learning
(49:10) The Art of the Jailbreak
(49:49) Unprompted Attention
(50:44) Get Involved
(51:33) Introducing
(51:52) In Other AI News
(52:45) Show Me the Money
(57:08) Nvidia Sells Out
(01:03:14) Quiet Speculations
(01:06:16) The Quest for Sane Regulations
(01:18:18) The Week in Audio
(01:20:13) Rhetorical Innovation
(01:34:29) Board of Anthropic
(01:37:08) Misaligned!
(01:39:22) Aligning a Smarter Than Human Intelligence is Difficult
(01:40:21) Americans Do Not Like AI
(01:42:37) People Are Worried About AI Killing Everyone
(01:44:09) Other People Are Not As Worried About AI Killing Everyone
(01:46:01) The Lighter Side
---
First published:
Source:
---
Narrated by TYPE III AUDIO.
---
Images from the article:
Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.
The big news of this week was of course the release of Claude 4 Opus. I offered two review posts: One on safety and alignment, and one on mundane utility, and a bonus fun post on Google's Veo 3.
I am once again defaulting to Claude for most of my LLM needs, although I often will also check o3 and perhaps Gemini 2.5 Pro.
On the safety and alignment front, Anthropic did extensive testing, and reported that testing in an exhaustive model card. A lot of people got very upset to learn that Opus could, if pushed too hard in the wrong situations engineered for these results, do things like report your highly unethical actions to authorities or try to blackmail developers into not being shut down or replaced. It is good that we now know about these things, and it was quickly observed that similar behaviors [...]
---
Outline:
(01:23) Language Models Offer Mundane Utility
(08:54) Now With Extra Glaze
(15:54) Get My Agent On The Line
(17:03) Language Models Don't Offer Mundane Utility
(22:49) Huh, Upgrades
(26:42) On Your Marks
(27:35) Choose Your Fighter
(33:40) Deepfaketown and Botpocalypse Soon
(37:51) Fun With Media Generation
(38:21) Playing The Training Data Game
(38:38) They Took Our Jobs
(46:51) The Art of Learning
(49:10) The Art of the Jailbreak
(49:49) Unprompted Attention
(50:44) Get Involved
(51:33) Introducing
(51:52) In Other AI News
(52:45) Show Me the Money
(57:08) Nvidia Sells Out
(01:03:14) Quiet Speculations
(01:06:16) The Quest for Sane Regulations
(01:18:18) The Week in Audio
(01:20:13) Rhetorical Innovation
(01:34:29) Board of Anthropic
(01:37:08) Misaligned!
(01:39:22) Aligning a Smarter Than Human Intelligence is Difficult
(01:40:21) Americans Do Not Like AI
(01:42:37) People Are Worried About AI Killing Everyone
(01:44:09) Other People Are Not As Worried About AI Killing Everyone
(01:46:01) The Lighter Side
---
First published:
Source:
---
Narrated by TYPE III AUDIO.
---
Images from the article:
Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.