LessWrong posts by zvi

By zvi

Audio narrations of LessWrong posts by zvi... more

Download on the App Store

Download on the App Store

Get it on Google Play

FAQs about LessWrong posts by zvi:

How many episodes does LessWrong posts by zvi have?

The podcast currently has 301 episodes available.

LessWrong posts by zvi episodes:

April 18, 2025 “o3 Will Use Its Tools For You” by Zvi
OpenAI has finally introduced us to the full o3 along with o4-mini.
Greg Brockman (OpenAI): Just released o3 and o4-mini! These models feel incredibly smart.
We’ve heard from top scientists that they produce useful novel ideas.
Excited to see their positive impact on people's daily lives and humanity's hardest problems!
Sam Altman: we expect to release o3-pro to the pro tier in a few weeks
By all accounts, this upgrade is a big deal. They are giving us a modestly more intelligent model, but more importantly giving it better access to tools and ability to discern when to use them, to help get more practical value out of it. The tool use, and the ability to string it together and persist, is where o3 shines.

The highest praise I can give o3 is that this was by far the most a model has been used as part of writing its [...]
---
Outline:
(01:56) What's In a Name
(02:51) My Current Model Use Heuristics
(04:21) Huh, Upgrades
(05:31) Use All the Tools
(09:47) Search the Web
(10:27) On Your Marks
(18:15) The System Prompt
(19:00) The o3 and o4-mini System Card
(23:17) Tests o3 Aced
(25:14) Hallucinations
(31:41) Instruction Hierarchy
(32:52) Image Refusals
(33:18) METR Evaluations for Task Duration and Misalignment
(42:45) Apollo Evaluations for Scheming and Deception
(44:40) We Are Insufficiently Worried About These Alignment Failures
(47:16) GPT-4.1 Also Has Some Issues
(50:08) Pattern Lab Evaluations for Cybersecurity
(51:45) Preparedness Framework Tests
(52:14) Biological and Chemical Risks (4.2)
(58:20) Cybersecurity (4.3)
(59:27) AI Self-Improvement (4.4)
(01:00:51) Perpetual Shilling
(01:01:54) High Praise
(01:09:31) Syncopathy
(01:11:58) Mundane Utility Versus Capability Watch
(01:16:33) o3 Offers Mundane Utility
(01:24:10) o3 Doesn't Offer Mundane Utility
(01:30:54) o4-mini Also Exists
(01:31:31) Colin Fraser Dumb Model Watch
(01:32:52) o3 as Forecaster
(01:34:31) Is This AGI?
---

First published:
April 18th, 2025

Source:
https://www.lesswrong.com/posts/u58AyZziQRAcbhTxd/o3-will-use-its-tools-for-you
---

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.
...more
1h 42min
April 17, 2025 “AI #112: Release the Everything” by Zvi
OpenAI has upgraded its entire suite of models. By all reports, they are back in the game for more than images.
GPT-4.1 and especially GPT-4.1-mini are their new API non-reasoning models. All reports are that GPT-4.1-mini especially is very good.
o3 is the new top of the line ChatGPT reasoning model, with o3-pro coming in a few weeks. Reports are that it too looks very good, even without us yet taking much advantage of its tool usage. If you have access, check it out. Full coverage is coming soon. There's also o4-mini and o4-mini-high.
Oh, they also made ChatGPT memory cover all your conversations, if you opt in, and gave us a version of Claude Code called Codex. And an update to their preparedness framework that I haven’t had time to examine yet.

Anthropic gave us (read-only for now) Google integration (as in GMail and Calendar to complement Drive), and [...]
---
Outline:
(01:32) Language Models Offer Mundane Utility
(03:08) Language Models Don't Offer Mundane Utility
(06:07) Huh, Upgrades
(09:57) On Your Marks
(12:42) Research Quickly, There's No Time
(15:16) Choose Your Fighter
(16:07) Deepfaketown and Botpocalypse Soon
(16:30) The Art of the Jailbreak
(16:44) Get Involved
(18:46) Introducing
(21:21) In Other AI News
(23:16) Come on OpenAI, Again?
(25:58) Show Me the Money
(26:41) In Memory Of
(28:28) Quiet Speculations
(32:17) America Restricts H20 Sales
(39:04) House Select Committee Report on DeepSeek
(48:46) Tariff Policy Continues To Be How America Loses
(58:53) The Quest for Sane Regulations
(01:01:03) The Week in Audio
(01:06:13) Rhetorical Innovation
(01:10:39) Aligning a Smarter Than Human Intelligence is Difficult
(01:14:40) AI 2027
(01:15:13) People Are Worried About AI Killing Everyone
(01:19:54) The Lighter Side
---

First published:
April 17th, 2025

Source:
https://www.lesswrong.com/posts/nycc4QxQAMkzmXmfz/ai-112-release-the-everything
---

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.
...more
1h 23min
April 16, 2025 “GPT-4.1 Is a Mini Upgrade” by Zvi
Yesterday's news alert, nevertheless: The verdict is in. GPT-4.1-Mini in particular is an excellent practical model, offering strong performance at a good price. The full GPT-4.1 is an upgrade to OpenAI's more expensive API offerings, it is modestly better but costs 5x as much. Both are worth considering for coding and various other API uses. If you have an agent or other app, it's at least worth trying plugging these in and seeing how they do.
This post does not cover OpenAI's new reasoning models. That was today's announcement, which will be covered in full in a few days, once we know more.
Introducing GPT-4.1, 4.1-mini and 4.1-nano
That's right, 4.1.
Here is their livestream, in case you aren’t like me and want to watch it.
On the one hand, I love that they might finally use a real version number with 4.1.
[...]

---
Outline:
(00:51) Introducing GPT-4.1, 4.1-mini and 4.1-nano
(03:59) On Your Marks
(06:41) Other Benchmarks
(13:55) Reactions
(17:55) Justice for GPT-4.5
(19:28) Safety Third
---

First published:
April 16th, 2025

Source:
https://www.lesswrong.com/posts/t5dMAHrmdXqNaNB6v/gpt-4-1-is-a-mini-upgrade
---

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.
...more
22min
April 15, 2025 “OpenAI #13: Altman at TED and OpenAI Cutting Corners on Safety Testing” by Zvi
Three big OpenAI news items this week were the FT article describing the cutting of corners on safety testing, the OpenAI former employee amicus brief, and Altman's very good TED Interview.
The FT detailed OpenAI's recent dramatic cutting back on the time and resources allocated to safety testing of its models.
In the interview, Chris Anderson made an unusually strong effort to ask good questions and push through attempts to dodge answering. Altman did a mix of giving a lot of substantive content in some places while dodging answering in others. Where he chose to do which was, itself, enlightening. I felt I learned a lot about where his head is at and how he thinks about key questions now.

The amicus brief backed up that OpenAI's current actions are in contradiction to the statements OpenAI made to its early employees.
There are also a few other related developments.
What [...]
---
Outline:
(01:26) The GPUs Are Melting
(02:36) On OpenAI's Ostensive Open Model
(03:25) Other People Are Not Worried About AI Killing Everyone
(04:18) What Even is AGI?
(05:19) The OpenAI AI Action Plan
(05:57) Copyright Confrontation
(07:43) The Ring of Power
(11:55) Safety Perspectives
(14:00) Autonomous Killer Robots
(15:15) Amicus Brief
(17:25) OpenAI Recklessly Races to Release
---

First published:
April 15th, 2025

Source:
https://www.lesswrong.com/posts/nnzb7prw2XC8cFhnP/openai-13-altman-at-ted-and-openai-cutting-corners-on-safety
---

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.
...more
23min
April 14, 2025 “Monthly Roundup #29: April 2025” by Zvi
In Monthly Roundup #28 I made clear I intend to leave the Trump administration out of my monthly roundups, for both better and worse, outside of my focus areas. Again, this does not mean I don’t have a lot to say or that those questions don’t matter. It means you should not rely on me as your only source of news and I pick my battles.
They are not making this easy.
I am going to stick to my guns. Trade and trading very much inside my focus areas, but for economics roundups, and in extreme cases AI roundups. Besides, you don’t need me to tell you that tariffs not only impose immense economic costs but also fail to achieve their primary policy aims and foster political dysfunction along the way. That question should already be answered by my t-shift. I do have a word about things [...]

---
Outline:
(01:12) Bad News
(02:49) Antisocial Media
(04:47) Technology Advances
(10:26) Variously Effective Altruism
(10:44) Government Working
(12:03) Jones Act Watch
(17:15) While I Cannot Condone This
(22:22) Architectural Musings
(22:26) A cool note from Christopher Alexander, I'm still a little bummed I never got to properly review A Pattern Language and it's probably too late now.
(23:49) Quickly, There's No Time
(25:14) Don't Sell Your Soul, You Won't Get Paid
(26:38) What To Do Instead
(28:18) Good News, Everyone
(30:01) We're Elite, You're Not
(33:02) Enjoy It While It Lasts
(34:44) For Your Entertainment
(37:16) An Economist Gets Lunch
(38:04) I Was Promised Flying Self-Driving Cars and Supersonic Jets
(42:22) Gamers Gonna Game Game Game Game Game
(45:05) Sports Go Sports
(45:52) The Lighter Side
---

First published:
April 14th, 2025

Source:
https://www.lesswrong.com/posts/Rwqsx4ktn8WBWaf4y/monthly-roundup-29-april-2025
---

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.
...more
51min
April 11, 2025 “On Google’s Safety Plan” by Zvi
Google Lays Out Its Safety Plans
I want to start off by reiterating kudos to Google for actually laying out its safety plan. No matter how good the plan, it's much better to write down and share the plan than it is to not share the plan, which in turn is much better than not having a formal plan.
They offer us a blog post, a full monster 145 page paper (so big you have to use Gemini!) and start off the paper with a 10 page summary.
The full paper is full of detail about what they think and plan, why they think and plan it, answers to objections and robust discussions. I can offer critiques, but I couldn’t have produced this document in any sane amount of time, and I will be skipping over a lot of interesting things in the full paper because [...]

---
Outline:
(00:58) Core Assumptions
(08:14) The Four Questions
(11:46) Taking This Seriously
(16:04) A Problem For Future Earth
(17:53) That's Not My Department
(23:22) Terms of Misuse
(27:50) Misaligned!
(36:17) Aligning a Smarter Than Human Intelligence is Difficult
(51:48) What Is The Goal?
(54:43) Have You Tried Not Trying?
(58:07) Put It To The Test
(59:44) Mistakes Will Be Made
(01:00:35) Then You Have Structural Risks
---

First published:
April 11th, 2025

Source:
https://www.lesswrong.com/posts/hvEikwtsbf6zaXG2s/on-google-s-safety-plan
---

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.
...more
1h 4min
April 10, 2025 “AI #111: Giving Us Pause” by Zvi
Events in AI don’t stop merely because of a trade war, partially paused or otherwise.
Indeed, the decision to not restrict export of H20 chips to China could end up being one of the most important government actions that happened this week. A lot of people are quite boggled about how America could so totally fumble the ball on this particular front, especially given what else is going on. Thus, I am going to put these issues up top again this week, with the hopes that we don’t have to do that again.
This week's previous posts covered AI 2027, both the excellent podcast about it and other people's response, and the release of Llama-4 Scout and Llama-4 Maverick. Both Llama-4 models were deeply disappointing even with my low expectations.
Upgrades continue, as Gemini 2.5 now powers Deep Research, and both it and Grok 3 are [...]

---
Outline:
(01:20) The Tariffs And Selling China AI Chips Are How America Loses
(08:37) Language Models Offer Mundane Utility
(13:13) Language Models Don't Offer Mundane Utility
(16:16) Huh, Upgrades
(20:50) Choose Your Fighter
(23:33) On Your Marks
(28:37) Deepfaketown and Botpocalypse Soon
(29:25) Fun With Image Generation
(33:37) Copyright Confrontation
(34:24) They Took Our Jobs
(39:24) Get Involved
(40:50) Introducing
(42:49) In Other AI News
(47:13) Show Me the Money
(50:41) Quiet Speculations
(53:35) The Quest for Sane Regulations
(57:35) The Week in Audio
(58:22) AI 2027
(01:02:31) Rhetorical Innovation
(01:05:23) Aligning a Smarter Than Human Intelligence is Difficult
---

First published:
April 10th, 2025

Source:
https://www.lesswrong.com/posts/RT2zbHdjYwuKXdHjB/ai-111-giving-us-pause
---

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.
...more
1h 10min
April 09, 2025 “Llama Does Not Look Good 4 Anything” by Zvi
Llama Scout (17B active parameters, 16 experts, 109B total) and Llama Maverick (17B active parameters, 128 experts, 400B total), released on Saturday, look deeply disappointing. They are disappointing on the level of ‘people think they have to be misconfigured to be this bad,’ and people wondering and debating how aggressively the benchmarks were gamed.
This was by far the most negative reaction I have seen to a model release, the opposite of the reaction to Gemini 2.5 Pro. I have seen similarly deeply disappointing and misleading releases, but they were non-American models from labs whose benchmarks and claims we have learned not to take as representing model capabilities.
After this release, I am placing Meta in that category of AI labs whose pronouncements about model capabilities are not to be trusted, that cannot be relied upon to follow industry norms, and which are clearly not on the [...]

---
Outline:
(01:31) Llama We Doing This Again
(03:05) Llama the License Favors Bad Actors
(04:35) Llama You Do It This Way
(09:35) Llama Fight in the Arena
(13:52) Llama Would You Cheat on Other Benchmarks
(19:25) Llama It So Bad on Independent Benchmarks
(27:55) Llama You Don't Like It
(32:30) Llama Should We Care
---

First published:
April 9th, 2025

Source:
https://www.lesswrong.com/posts/FHMcHmW4RKY5JiArJ/llama-does-not-look-good-4-anything
---

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.
...more
34min
April 08, 2025 “AI 2027: Responses” by Zvi
Yesterday I covered Dwarkesh Patel's excellent podcast coverage of AI 2027 with Daniel Kokotajlo and Scott Alexander. Today covers the reactions of others.
Kevin Roose in The New York Times
Kevin Roose covered Scenario 2027 in The New York Times.
Kevin Roose: I wrote about the newest AGI manifesto in town, a wild future scenario put together by ex-OpenAI researcher @DKokotajlo and co.
I have doubts about specifics, but it's worth considering how radically different things would look if even some of this happened.
Daniel Kokotajlo: AI companies claim they’ll have superintelligence soon. Most journalists understandably dismiss it as hype. But it's not just hype; plenty of non-CoI’d people make similar predictions, and the more you read about the trendlines the more plausible it looks. Thank you & the NYT!

The final conclusion is supportive of this kind of work, and Kevin points out that expectations at the major [...]
---
Outline:
(00:21) Kevin Roose in The New York Times
(02:56) Eli Lifland Offers Takeaways
(04:23) Scott Alexander Offers Takeaways
(05:34) Others Takes on Scenario 2027
(05:39) Having a Concrete Scenario is Helpful
(08:37) Writing It Down Is Valuable Even If It Is Wrong
(10:00) Saffron Huang Worries About Self-Fulfilling Prophecy
(18:18) Phillip Tetlock Calibrates His Skepticism
(21:38) Jan Kulveit Wants to Bet
(23:08) Matthew Barnett Debates How To Evaluate the Results
(24:38) Teortaxes for China and Open Models and My Response
(31:53) Others Wonder About PRC Passivity
(33:40) Timothy Lee Remains Skeptical
(35:16) David Shapiro for the Accelerationists and Scott's Response
(45:29) LessWrong Weighs In
(46:59) Other Reactions
(50:05) Next Steps
(52:34) The Lighter Side
---

First published:
April 8th, 2025

Source:
https://www.lesswrong.com/posts/gyT8sYdXch5RWdpjx/ai-2027-responses
---

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.
...more
54min
April 07, 2025 “AI 2027: Dwarkesh’s Podcast with Daniel Kokotajlo and Scott Alexander” by Zvi
Daniel Kokotajlo has launched AI 2027, Scott Alexander introduces it here. AI 2027 is a serious attempt to write down what the future holds. His ‘What 2026 Looks Like’ was very concrete and specific, and has proved remarkably accurate given the difficulty level of such predictions.
I’ve had the opportunity to play the wargame version of the scenario described in 2027, and I reviewed the website prior to publication and offered some minor notes. Whenever I refer to a ‘scenario’ in this post I’m talking about Scenario 2027.
There's tons of detail here. The research here, and the supporting evidence and citations and explanations, blow everything out of the water. It's vastly more than we usually see, and dramatically different from saying ‘oh I expect AGI in 2027’ or giving a timeline number. This lets us look at what happens in concrete detail, figure out where we [...]

---
Outline:
(02:00) The Structure of These Post
(03:37) Coverage of the Podcast
---

First published:
April 7th, 2025

Source:
https://www.lesswrong.com/posts/vnkH6JrGu2AxtDTyu/ai-2027-dwarkesh-s-podcast-with-daniel-kokotajlo-and-scott
---

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.
...more
49min

FAQs about LessWrong posts by zvi:

How many episodes does LessWrong posts by zvi have?

The podcast currently has 301 episodes available.