If you are Meta, and you want to attract top AI talent, you have a problem, because no one wants to work for you or on your products. So it is going to cost you. Mark Zuckerberg has decided he will pay what it takes to get at least some very good talent.

If you are the rest of us, especially someone seeking an entry-level job and for whom $100 million signing bonuses are not flooding your inboxes, things are getting rougher. AI might not yet be destroyed a lot of jobs, but it is doing a number on the job application process.

There's a lot of other stuff going on as per usual.

Anthropic won a case establishing that model training is fair use.

Yesterday I shared various Tales of Agentic Misalignment, and we have some more related fun since then that is included [...]

---

Outline:

(01:20) Language Models Offer Mundane Utility

(08:16) Language Models Don't Offer Mundane Utility

(10:49) Huh, Upgrades

(11:29) On Your Marks

(12:01) Choose Your Fighter

(13:58) Deepfaketown and Botpocalypse Soon

(16:56) Release the Hounds

(19:31) Copyright Confrontation

(21:12) Cheaters Gonna Cheat (x5)

(23:39) Fun With Media Generation

(24:07) Get My Agent On The Line

(24:36) They Took Our Jobs

(27:24) Get Involved

(28:12) Introducing

(30:40) In Other AI News

(32:59) Gemini Sings The Blues

(34:12) Show Me the Money

(38:23) Quiet Speculations

(42:14) Timelines

(43:18) Jack Clark Testifies

(52:52) The Quest for Sane Regulations

(55:57) Chip City

(56:08) The Week in Audio

(57:36) Rhetorical Innovation

(01:05:09) Be Prepared

(01:05:42) Misaligned!

(01:09:10) Aligning a Smarter Than Human Intelligence is Difficult

(01:14:58) Other People Are Not As Worried About AI Killing Everyone

(01:15:36) 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.

(Note: This is NOT being posted on my Substack or Wordpress, but I do want a record of it that is timestamped and accessible for various reasons, so I'm putting this here, but do not in any way feel like you need to read it unless it sounds like fun to you.)

We all deserve some fun. This is some of mine.

I will always be a gamer and a sports fan, and especially a game designer, at heart. The best game out there, the best sport out there is Love Island USA, and I’m getting the joy of experiencing it in real time.

A Brief Ode To One Of Our Beloved Games

Make no mistake. This is a game, a competition, and the prizes are many.

The people agree. It is more popular than ever, with huge social media engagement.

I got into this through the ‘horrible people [...]

---

Outline:

(00:45) A Brief Ode To One Of Our Beloved Games

(04:03) Nice Hat

(05:12) The Story of S7 Casa Amor So Far In Brief

(08:45) This Was Very Obviously On Purpose

(12:09) The Problem With Casa Amor Is The Incentives

(14:51) What Do We Get Instead?

(16:43) Return To The Villa

(19:51) Was That A Good Idea? Was It Done Honestly?

---

First published:

Source:

---

Narrated by TYPE III AUDIO.

There was what everyone agrees was a high quality critique of the timelines component of AI 2027, by the LessWrong user and Substack writer Titotal.

It is great to have thoughtful critiques like this. The way you get actual thoughtful critiques like this, of course, is to post the wrong answer (at length) on the internet, and then respond by listening to the feedback and by making your model less wrong.

This is a high-effort, highly detailed, real engagement on this section, including giving the original authors opportunity to critique the critique, and warnings to beware errors, give time to respond, shares the code used to generate the graphs, engages in detail, does a bunch of math work, and so on. That is The Way.

So, Titotal: Thank you.

I note up front that at least Daniel Kokotajlo has indeed adjusted his estimates, and has [...]

---

Outline:

(02:00) The Headline Message Is Not Ideal

(09:19) An Explanation of Where Superexponentiality Is Coming From

(10:24) Three Methods

(10:53) Time Horizon Extension Method

(17:53) The Public Versus Internal Gap

(19:41) The Difficulty Gap

(21:38) Recent Progress

(23:36) Infinite Time Horizons

(24:53) Intermediate Speedups

(28:02) Is There A Flawed Graph Still Up?

(29:07) Some Skepticism About Projection

(29:58) Part 2: Benchmarks and Gaps and Beyond

(30:03) Benchmarks

(35:16) The Time Horizon Part of the Second Model

(37:44) Why The Thresholds?

(39:14) The Gap Model

(39:42) Eli Responds On LessWrong

(45:10) On Eli's Recent Update

(46:09) Conclusion

(50:38) Perhaps The Most Important Disagreement

The original text contained 1 footnote which was omitted from this narration.

---

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.

Edition #9, that School is Hell, turned out to hit quite the nerve.

Thus, I’m going to continue with the system of making the roundups have more focus in their themes, with this one being the opposite of school questions, except for the question of banning phones in schools which seemed to fit.

Table of Contents

Metal Health

Henry Shevlin: I asked a high school teacher friend about the biggest change in teens over the past decade. His answer was [...]

---

Outline:

(00:28) Metal Health

(02:02) Coercion

(03:18) Game Theoretically Sound Discipline

(05:38) The Joy of Doing Nothing

(07:31) ADHD Exists But So Do Boys

(08:32) Sports Go Sports

(10:22) On the Big Screen

(12:41) Kids Media Is Often Anti-Capitalist Propaganda

(13:45) Culture

(15:23) Travel

(16:25) Phone a Friend

(20:11) The Case For Phones

(26:07) Ban Cell Phones in Schools

(28:41) A Sobering Thought

---

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.

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:

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.

That's right. I said Part 1. The acceleration continues.

I do not intend to let this be a regular thing. I will (once again!) be raising the bar for what gets included going forward to prevent that. But for now, we’ve hit my soft limit, so I’m splitting things in two, mostly by traditional order but there are a few things, especially some videos, that I’m hoping to get to properly before tomorrow, and also I’m considering spinning out my coverage of The OpenAI Files.

Tomorrow in Part 2 we’ll deal with, among other things, several new videos, various policy disputes and misalignment fun that includes the rising number of people being driven crazy.

Table of Contents

---

Outline:

(00:53) Language Models Offer Mundane Utility

(03:12) Language Models Don't Offer Mundane Utility

(09:53) Humans Do Not Offer Mundane Utility

(10:59) Langage Models Should Remain Available

(12:51) Get My Agent On The Line

(15:45) Have My Agent Call Their Agent

(18:09) Beware Prompt Injections

(21:13) Unprompted Attention

(23:18) Huh, Upgrades

(26:02) Memories

(27:56) Cheaters Gonna Cheat Cheat Cheat Cheat Cheat

(36:40) On Your Marks

(40:05) Fun With Media Generation

(41:20) Copyright Confrontation

(45:13) Deepfaketown and Botpocalypse Soon

(50:28) Liar Liar

(52:25) They Took Our Jobs

(54:18) No, Not Those Jobs

(58:37) All The Jobs Everywhere All At Once

(01:04:44) The Void

(01:05:58) Into the Void

(01:06:49) The Art of the Jailbreak

(01:07:18) Get Involved

(01:07:32) Introducing

(01:07:55) In Other AI News

(01:13:26) Show Me the Money

---

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.

Google recently came out with Gemini-2.5-0605, to replace Gemini-2.5-0506, because I mean at this point it has to be the companies intentionally fucking with us, right?

Google: Our updated Gemini 2.5 Pro Preview continues to excel at coding, helping you build more complex web apps. We’ve also added thinking budgets for more control over cost and latency. GA is coming in a couple of weeks…

We’re excited about this latest model and its improved performance. Start building with our new preview as support for the 05-06 preview ends June 19th.

Sundar Pichai (CEO Google): Our latest Gemini 2.5 Pro update is now in preview.

It's better at coding, reasoning, science + math, shows improved performance across key benchmarks (AIDER Polyglot, GPQA, HLE to name a few), and leads @lmarena_ai with a 24pt Elo score jump since the previous version.

We also heard your feedback [...]

---

Outline:

(02:06) What's In A Name

(02:49) On Your Marks

(08:05) Gemini 2.5 Flash Lite Preview

(08:24) The Model Report

(11:33) Safety Testing

(15:17) General Reactions to Gemini 2.5-Pro 0605

(16:48) The Personality Upgrades Are Bad

(19:36) 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.

You can now have o3 throw vastly more compute at a given problem. That's o3-pro.

Should you have o3 throw vastly more compute at a given problem, if you are paying the $200/month subscription price for ChatGPT Pro? Should you pay the $200, or the order of magnitude markup over o3 to use o3-pro in the API?

That's trickier. Sometimes yes. Sometimes no. My experience so far is that waiting a long time is annoying, sufficiently annoying that you often won’t want to wait. Whenever I ask o3-pro something, I often also have been asking o3 and Opus.

Using the API at scale seems prohibitively expensive for what you get, and you can (and should) instead run parallel queries using the chat interface.

The o3-pro answers have so far definitely been better than o3, but the wait is usually enough to break my workflow and [...]

---

Outline:

(03:07) o3-Pro Versus o3

(06:07) Give Me a Minute

(07:33) On Your Marks

(10:04) Reactions

(23:30) Brevity Is The Soul Of Wit

(24:08) Where Art Thou O System Card

(27:05) The Price Drop on o3

---

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.

Overall Take

---

Outline:

(00:45) Overall Take

(03:04) RTFB: The RAISE Act

(03:26) Definitions: Frontier Model

(07:24) Definitions: Critical Harm

(08:59) Definition: Safety Incident

(10:22) You Have To Report Safety Incidents The Way We Report Cybersecurity Breaches

(11:31) Definition: Safety and Security Protocol (SSP)

(12:31) You Have To Have And Implement An SSP

(13:44) You Have To Not Release Models That Would Create An Unreasonable Risk Of Critical Harm

(13:59) You Shall Not Knowingly Make False Or Materially Misleading Statements or Omissions

(14:10) What Are The Consequences?

---

First published:

Source:

---

Narrated by TYPE III AUDIO.

It's always a nice break to see what else is going on out there.

Bad News

Study finds sleep in male full-time workers falls as income rises, with one cause being other leisure activities substituting for sleep. It makes sense that sleep doesn’t cost money while other things often do, but the marginal cost of much leisure is very low. I don’t buy this as the cause. Perhaps reverse causation, those who need or prefer less sleep earn more money?

The productivity statistics continue to be awful, contra Alex Tabarrok part of this recent -3.88% Q1 print is presumably imports anticipating tariffs driving down measured GDP and thus productivity. The more I wonder what's wrong with the productivity statistics the more I think they’re just a terrible measure of productivity?

A model of America's electoral system that primary voters don’t know much about candidate positions [...]

---

Outline:

(00:16) Bad News

(01:58) Rules of Confidentiality

(02:44) Something Is Wrong On The Internet

(06:21) Government Working

(08:39) Jones Act Watch

(09:05) Signatures as Solemnization

(12:12) Antisocial Media

(19:43) Technology Advances

(25:12) For Science!

(25:37) Variously Effective Altruism

(28:44) Lab Grown Meat Ban Redux

(36:04) Opportunity Knocks

(36:16) Good Books

(36:57) Commitment

(39:14) How To Pass On Generational Wealth

(42:21) Rain On Your Wedding Day

(44:14) Social Versus Transactional Exchanges

(48:24) While I Cannot Condone This

(54:07) Costly Signaling Is a Weird Name

(56:58) Travel en-US-AvaMultilingualNeural__ Trademark symbol (™) shown as large dark letters

(01:00:22) Good News, Everyone

(01:04:28) Put On a Happy Face

(01:06:44) Egg Prices Keep Going Up

(01:09:04) Cultural Stagnation

(01:11:54) For Your Entertainment

(01:22:33) Gamers Gonna Game Game Game Game Game

(01:27:00) I Was Promised Flying Self-Driving Cars

(01:31:06) 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.