Start / LessWrong (30+ Karma) / Ai 121 part 2 the openai files by zvi

LessWrong (30+ Karma)

“AI #121 Part 2: The OpenAI Files” by Zvi

78 min • 21 juni 2025

You can find Part 1 here. This resumes the weekly, already in progress. The primary focus here is on the future, including policy and alignment, but also the other stuff typically in the back half like audio, and more near term issues like ChatGPT driving an increasing number of people crazy.

If you haven’t been following the full OpenAI saga, the OpenAI Files will contain a lot of new information that you really should check out. If you’ve been following, some of it will likely still surprise you, and help fill in the overall picture behind the scenes to match the crazy happening elsewhere.

At the end, we have some crazy new endorsements for Eliezer Yudkowsky's upcoming book, If Anyone Builds It, Everyone Dies. Preorders make a difference in helping the book get better reach, and I think that will help us all have a much better [...]

---

Outline:

(01:02) Cheaters Gonna Cheat Cheat Cheat Cheat Cheat

(02:12) Quiet Speculations

(07:44) Get Involved

(07:53) Thinking Machines

(08:04) California Reports

(10:39) The Quest for Sane Regulations

(13:40) What Is Musk Thinking?

(16:54) Why Do We Care About The 'AI Race'?

(23:52) Chip City

(26:29) Pick Up The Phone

(26:46) The OpenAI Files

(33:13) The Week in Audio

(38:03) Rhetorical Innovation

(42:37) Aligning a Smarter Than Human Intelligence is Difficult

(43:33) Misaligned!

(45:28) Emergently Misaligned!

(47:17) ChatGPT Can Drive People Crazy

(57:27) Misalignment By Default

(01:04:59) People Are Worried About AI Killing Everyone

(01:06:46) Other People Are Not As Worried About AI Killing Everyone

(01:07:04) The Too Open Model

(01:13:25) A Good Book

(01:17:12) The Lighter Side

---

First published:
June 20th, 2025

Source:
https://www.lesswrong.com/posts/oHQ6FdcNymCt2rqbi/ai-121-part-2-the-openai-files

---

Narrated by TYPE III AUDIO.

---

Images from the article:

Social media chat about privacy concerns between a user and Meta AI.
Social media conversation about demographics and dating preferences between two users.
Social media post showing advice about browser history and relationship communication.
This image appears to be a political statement regarding technology export controls, featuring text that discusses concerns about AI chip technology and China's access to it. The quote discusses the CEOs of major tech companies (Intel, Nvidia, and Qualcomm) and their lobbying efforts regarding export restrictions. The statement expresses concern about national security implications of advanced AI chip access.
A humorous chat exchange where someone quickly regrets sharing their phone number.
Timeline showing AI companies' valuations and leadership concerns from 2021-2025. Companies are Anthropic ($62B), xAI ($50B), SSI ($32B), and TM ($10B), with quoted concerns about leadership and control.
The image appears to be an AI-generated output based on a text prompt shown at the bottom. The generated image shows two individuals standing near an older model truck with some rust, in front of a house with a white picket fence. One person is wearing a bright yellow dress, while the other is seated in a wheelchair, wearing casual attire including a cap and brown jacket. The lighting suggests it was taken during
A cartoon meme comparing different perspectives on AI planning, showing contrasting views between most people and accelerationists, depicted inside what appears to be a bus or train with scenic mountain views visible through the windows.
This appears to be a social media post discussing a retiree's potential relocation to Medellin, Colombia. The person describes their financial situation, health conditions, and desire to find companionship while living on a fixed income of $1,900 monthly. The response provides information about living costs in Medellin, noting that apartments in good neighborhoods can range from $300-500 per month. The post illustrates a common scenario of retirees considering international retirement destinations where their fixed income might provide a more comfortable lifestyle. However, I should point out some concerning aspects of this post. The person's stated goal of finding
pdx1 tweets:
This appears to be an AI chat interaction where a user asks several questionable driving-related queries. The AI provides professional, safety-focused responses emphasizing that distracted driving is dangerous and illegal, explaining how any activities that divert attention from driving can lead to accidents. The exchange includes three back-and-forth messages, with the user asking increasingly unusual questions about specific driving distractions while the AI maintains a consistent, responsible tone focused on road safety and responsible vehicle operation.

Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

Senaste avsnitt

“Vitalik’s Response to AI 2027” by Daniel Kokotajlo

12 juli | 24 min

“the jackpot age” by thiccythot

12 juli | 13 min

“Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity” by habryka

11 juli | 12 min

[Linkpost] “Guide to Redwood’s writing” by Julian Stastny

11 juli | 1 min

“So You Think You’ve Awoken ChatGPT” by JustisMills

11 juli | 18 min
Vad är en podd?

En liten tjänst av I'm With Friends. Finns även på engelska.

Podcastbild

00:00 -00:00
00:00 -00:00