It has been brutal out there for someone on my beat. Everyone extremely hostile, even more than usual. Extreme positions taken, asserted as if obviously true. Not symmetrically, but from all sides nonetheless. Constant assertions of what happened in the last two weeks that are, as far as I can tell, flat out wrong, largely the result of a well-implemented media campaign. Repeating flawed logic more often and louder.

The bright spot was offered by Vitalik Buterin, who offers a piece entitled ‘My techo–optimism,’ proposing what he calls d/acc for defensive (or decentralized, or differential) accelerationism. He brings enough nuance and careful thinking, and clear statements about existential risk and various troubles ahead, to get strong positive reactions from the worried. He brings enough credibility and track record, and enough shibboleths, to get strong endorsements from the e/acc crowd, despite his acknowledgement of existential risk and the dangers [...]

---

Outline:

(02:06) Language Models Offer Mundane Utility

(03:52) Language Models Don’t Offer Mundane Utility

(07:37) Q Continuum

(13:04) OpenAI, Altman and Safety

(15:30) A Better Way to Do RLHF

(19:26) Fun with Image Generation

(19:34) Deepfaketown and Botpocalypse Soon

(19:49) They Took Our Jobs

(20:35) Get Involved

(22:21) Introducing

(22:39) In Other AI News

(25:10) It's a Who?

(28:22) What About E/Acc?

(31:54) Vitalik Offers His Version of Techno-Optimism

(40:47) Quiet Speculations

(44:38) AI Agent Future

(47:25) The Quest for Sane Regulations

(51:28) The Week in Audio

(56:13) Rhetorical Innovation

(59:06) Aligning a Smarter Than Human Intelligence is Difficult

(01:06:36) People Might Also Worry About AI Killing Only Some of Them

(01:07:37) People Are Worried About AI Killing Everyone

(01:08:43) Other People Are Not As Worried About AI Killing Everyone

(01:12:38) Please Speak Directly Into This Microphone

(01:13:54) The Lighter Side

---

First published:

Source:

Narrated by TYPE III AUDIO.