Pluralistic: Daily...
Pluralistic: The Post-American Internet (01 Jan 2026)
Today's links The Post-American Internet: My speech from Hamburg's Chaos Communications Congress....
1st Jan 2026
Today's links The Post-American Internet: My speech from Hamburg's Chaos Communications Congress. Hey look at this: Delights to delectate. Object permanence: Error code 451; Public email address Mansplaining Lolita; NSA backdoor in Juniper Networks; Don't bug out; Nurses whose...
Caitlin’s Newsletter
Israel Bans Aid Groups For The Same Reason It Bans Journalists, And Other Notes
Reading by Tim Foley:
1st Jan 2026
laike9m's blog
Avoid Mini-frameworks
没有简介, 请去看原文吧
24th Dec 2025
Max Woolf's Blog
Nano Banana Pro is the best AI image generator, with caveats
The problem with Nano Banana Pro is that it’s too good.
22nd Dec 2025
The problem with Nano Banana Pro is that it’s too good.
Max Woolf's Blog
Nano Banana Pro is the best AI image generator, with caveats
The problem with Nano Banana Pro is that it’s too good.
22nd Dec 2025
The problem with Nano Banana Pro is that it’s too good.
One Useful Thing
The Shape of AI: Jaggedness, Bottlenecks and Salients
And why Nano Banana Pro is such a big deal
20th Dec 2025
And why Nano Banana Pro is such a big deal
Mind Prison
AI Vibe Coding, Is It Working? No
Notes From the Desk: No. 53 - 2025.12.18
18th Dec 2025
Notes From the Desk: No. 53 - 2025.12.18
Eugene Yan
2025 Year in Review
An eventful year of progress in health and career, while making time for travel and reflection.
14th Dec 2025
An eventful year of progress in health and career, while making time for travel and reflection.
matt.sh
Opsmas 2025 Day 1: haproxy & mpreg
Opsmas 2025 Day 1: new haproxy config system & mpreg
TOC:
haproxy-translate
mpreg
merry opsmas...
12th Dec 2025
Opsmas 2025 Day 1: new haproxy config system & mpreg
TOC:
haproxy-translate
mpreg
merry opsmas 2025.
this year i’ve created (somehow) more code than ever before including both new projects and cleaning up existing/abandoned projects.
I’d like to announce/release about 20 new...
Armin Ronacher's...
Let’s Destroy The European Union!
Elon Musk is not
happy
with the EU fining his X platform and is currently on a tweet...
9th Dec 2025
Elon Musk is not
happy
with the EU fining his X platform and is currently on a tweet rampage
complaining about it. Among other things, he wants the whole EU to be
abolished. He sadly is hardly the first wealthy American to share their
opinions on European politics lately. I’m...
Artificial Ignorance
AI Roundup 147: OpenAI's Code Red
December 5, 2025.
5th Dec 2025
Xena
Formalization of Erdős problems
[This is a guest post by Boris Alexeev. Now over to Boris.] I’m here to tell you about various...
5th Dec 2025
[This is a guest post by Boris Alexeev. Now over to Boris.] I’m here to tell you about various exciting developments centering on Erdős problems, especially involving the formalization of old and new mathematics using artificial intelligence. Background As is … Continue reading →
seangoedecke.com RSS...
AI detection tools cannot prove that text is AI-generated
5th Dec 2025
Piotr Migdał's Blog
Outside of the bubble, AI is Black Mirror
5th Dec 2025
Weighty Thoughts
The Boring Phase of AI
Advancements Will Become Less Visible—But More Impactful
4th Dec 2025
Advancements Will Become Less Visible—But More Impactful
AI Tidbits
Getting the most out of Claude Code
How to successfully on board Claude as a new engineer on your team
3rd Dec 2025
How to successfully on board Claude as a new engineer on your team
Marcus on AI
OpenAI’s “Code Red”
Maybe people should have seen this coming?
3rd Dec 2025
Maybe people should have seen this coming?
fast.ai
Stop Saying Boredom is Good for Kids
Chronic boredom causes stress, disengagement, and poor well-being in adults. So why do we glorify it...
3rd Dec 2025
Chronic boredom causes stress, disengagement, and poor well-being in adults. So why do we glorify it for children?
Giles' blog
Writing an LLM from scratch, part 28 -- training a base model from scratch on an RTX 3090
Having worked through the main body of Sebastian Raschka's book
"Build a Large Language Model (from...
2nd Dec 2025
Having worked through the main body of Sebastian Raschka's book
"Build a Large Language Model (from Scratch)",
I wanted to try an experiment: is it possible to train a base model of my
own, on my own hardware?
The book shows you how to train your LLM, does a basic training run
on...
AI for Software...
AI for SWEs 73: What Ilya Saw and the Time of TPUs
Plus, the White House unifies US AI research and Antigravity exposes why agents are so hard to build
2nd Dec 2025
Plus, the White House unifies US AI research and Antigravity exposes why agents are so hard to build
fast.ai
Stop Saying Boredom is Good for Kids
Chronic boredom is harmful to adults, causing stress, disengagement, and poor well-being. Academic...
2nd Dec 2025
Chronic boredom is harmful to adults, causing stress, disengagement, and poor well-being. Academic researchers have shown that boredom in the workplace can be just as damaging as burnout. But search for information about childhood boredom and you’ll find the opposite message:...
Win Vector LLC
Text as a “Market for Lemons”
I would like to write a bit about text. That is: technical writing, legal briefs, or even an opinion...
2nd Dec 2025
I would like to write a bit about text. That is: technical writing, legal briefs, or even an opinion piece such as this note. Such writings make up much of our society and form a “marketplace of ideas.” Texts are now very cheap to produce using large language models (LLMs). […]
Strange Loop Canon
Contra Scott on AI safety and the race with China
2nd Dec 2025
Anil Dash
Vibe Coding: Empowering and Imprisoning
In case you haven’t been following the world of software development closely, it’s good to know that...
2nd Dec 2025
In case you haven’t been following the world of software development closely, it’s good to know that vibe coding — using LLM tools to assist with writing code — can help enable many people to create apps or software that they wouldn’t otherwise be able to make. This has led to an...
Pluralistic: Daily...
Pluralistic: Meta's new top EU regulator is contractually prohibited from hurting Meta's feelings...
Today's links Meta's new top EU regulator is contractually prohibited from saying mean things about...
1st Dec 2025
Today's links Meta's new top EU regulator is contractually prohibited from saying mean things about Meta: It's one thing to hire an ex-Meta lobbyist, another entirely if she's signed a non-disparagement contract. Hey look at this: Delights to delectate. Object permanence: Custom...
Don't Worry About...
Claude Opus 4.5 Is The Best Model Available
Claude Opus 4.5 is the best model currently available.
1st Dec 2025
Claude Opus 4.5 is the best model currently available.
IEEE Spectrum
The Next Frontier in AI Isn’t More Data
For the past decade, progress in artificial intelligence has been measured by scale: bigger models,...
1st Dec 2025
For the past decade, progress in artificial intelligence has been measured by scale: bigger models, larger datasets, and more compute. That approach delivered astonishing breakthroughs in large language models (LLMs); in just five years, AI has leapt from models like GPT-2, which...
Caitlin’s Newsletter
The Craziest Thing In The World Is That We Could End Poverty, But We Don't
Listen to a reading of this article (reading by Tim Foley):
1st Dec 2025
Listen to a reading of this article (reading by Tim Foley):
the singularity is...
Replacing my MacBook
I’ve been trying to replace my MacBook. I switched from an iPhone to a Z Fold 5 (and now a Z Fold 7)...
28th Nov 2025
I’ve been trying to replace my MacBook. I switched from an iPhone to a Z Fold 5 (and now a Z Fold 7) which went well, so I’ve been trying to do the same with my laptop.
For phones, this was easier because the Samsung hardware is already better. Sadly, there’s no better laptop...
Strange Loop Canon
Epicycles All The Way Down
“All models are wrong, but some are useful.” — George E.
26th Nov 2025
“All models are wrong, but some are useful.” — George E.
Piotr Migdał's Blog
Nano Banana Pro: raw intelligence with tool use
24th Nov 2025
Eugene Yan
Product Evals in Three Simple Steps
Label some data, align LLM-evaluators, and run the eval harness with each change.
23rd Nov 2025
Label some data, align LLM-evaluators, and run the eval harness with each change.
Win Vector LLC
The Fermi Paradox Just From Rarity
Sabine Hossenfelder’s video “Odds Are The Universe is Full Of Intelligent Life, Mathematician Finds”...
22nd Nov 2025
Sabine Hossenfelder’s video “Odds Are The Universe is Full Of Intelligent Life, Mathematician Finds” recently pointed out a new paper on the Fermi paradox: Antal Veres “The solitude zone: A probabilistic window for singular lifeform existence”, Acta Astronautica Volume 238, Part...
Rachel Thomas, PhD
A Decade of Writing Stuff That People, To My Surprise, Actually Read
As a former mathematician, I was used to nobody reading what I wrote. So when I first began blogging...
17th Nov 2025
As a former mathematician, I was used to nobody reading what I wrote. So when I first began blogging in 2015, I never expected that several of my blog posts would go viral or to have multiple journalists contact me (including from NPR, Wired, and Fortune), make the front page of...
Rachel Thomas, PhD
A Decade of Writing Stuff That People, To My Surprise, Actually Read
As a former mathematician, I was used to nobody reading what I wrote. So when I first began blogging...
17th Nov 2025
As a former mathematician, I was used to nobody reading what I wrote. So when I first began blogging in 2015, I never expected that several of my blog posts would go viral or to have multiple journalists contact me (including from NPR, Wired, and Fortune), make the front page of...
Max Woolf's Blog
Nano Banana can be prompt engineered for extremely nuanced AI image generation
Nano Banana allows 32,768 input tokens and I’m going to try to use them all dammit.
13th Nov 2025
Nano Banana allows 32,768 input tokens and I’m going to try to use them all dammit.
Max Woolf's Blog
Nano Banana can be prompt engineered for extremely nuanced AI image generation
Nano Banana allows 32,768 input tokens and I’m going to try to use them all dammit.
13th Nov 2025
Nano Banana allows 32,768 input tokens and I’m going to try to use them all dammit.
One Useful Thing
Giving your AI a Job Interview
As AI advice becomes more important, we are going to need to get better at assessing it
12th Nov 2025
As AI advice becomes more important, we are going to need to get better at assessing it
Machine Learning for...
How to Build Your First Recommendation System (Easy)
A step-by-step guide to training and serving a collaborative filtering model to serve users content
11th Nov 2025
A step-by-step guide to training and serving a collaborative filtering model to serve users content
AI for Software...
How to Build Your First Recommendation System (Easy)
A step-by-step guide to training and serving a collaborative filtering model to serve users content
11th Nov 2025
A step-by-step guide to training and serving a collaborative filtering model to serve users content
fast.ai
A Guide to Solveit Features
An overview of the features of the Solveit platform, which is designed to make exploration and...
7th Nov 2025
An overview of the features of the Solveit platform, which is designed to make exploration and iterative development easier and faster.
Dylan Huang Blog
The Untold System Design Problem in LLM Inference
7th Nov 2025
Artificial Ignorance
Is a CS Degree Still Worth It?
The junior developer crisis and the value of CS fundamentals.
6th Nov 2025
The junior developer crisis and the value of CS fundamentals.
Weighty Thoughts
Models Are Not Exponentially Growing in Size
Stating the Obvious (for AI Experts) That's Non-Obvious To Everyone Else
5th Nov 2025
Stating the Obvious (for AI Experts) That's Non-Obvious To Everyone Else
Arguing with...
"Flow" in software development is overrated
Tackling the myth that "flow" is critical for software development. Decided not to go with the...
5th Nov 2025
Tackling the myth that "flow" is critical for software development. Decided not to go with the obvious "Flow considered harmful" title. You're welcome.
Anil Dash
Turn the volume up.
Today marked a completely new moment for New York City, and for America. There will be countless...
5th Nov 2025
Today marked a completely new moment for New York City, and for America. There will be countless attempts at analysis and reflection and what-does-it-all-mean in the days to come, along with an unimaginable number of hateful attacks. But what's worth reflecting on right now is...
Don't Worry About...
OpenAI: The Battle of the Board: Ilya's Testimony
New Things Have Come To Light
4th Nov 2025
New Things Have Come To Light
Giles' blog
Writing an LLM from scratch, part 26 -- evaluating the fine-tuned model
This post is on the second half of chapter 7 of
Sebastian Raschka's book
"Build a Large Language...
3rd Nov 2025
This post is on the second half of chapter 7 of
Sebastian Raschka's book
"Build a Large Language Model (from Scratch)".
In the last post I covered
the part of the chapter that covers instruction fine-tuning; this time round,
we evaluate our model -- particularly interestingly, we...
Mind Prison
Grokipedia or Slopipedia? Is it Truthful and Accurate?
Notes From the Desk: No. 52 - 2025.11.02
3rd Nov 2025
Notes From the Desk: No. 52 - 2025.11.02
seangoedecke.com RSS...
Is it worrying that 95% of AI enterprise projects fail?
3rd Nov 2025
Armin Ronacher's...
Absurd Workflows: Durable Execution With Just Postgres
It’s probably no surprise to you that we’re building
agents
somewhere. Everybody does it. Building...
3rd Nov 2025
It’s probably no surprise to you that we’re building
agents
somewhere. Everybody does it. Building a good agent, however, brings back
some of the historic challenges involving durable execution.
Entirely unsurprisingly, a lot of people are now building durable...
Marcus on AI
Too big to fail?
Get ready for chaos
2nd Nov 2025
Pluralistic: Daily...
Pluralistic: There's one thing EVERY government can do to shrink Big Tech (01 Nov 2025)
Today's links There's one thing EVERY government can do to shrink Big Tech: The path to a...
1st Nov 2025
Today's links There's one thing EVERY government can do to shrink Big Tech: The path to a post-American internet. Hey look at this: Delights to delectate. Object permanence: D2020; Sony rootkit; Public Enemy vs the internet; NYC plute Hallowe'en. Upcoming appearances: Where to...
Caitlin’s Newsletter
New Painting: Roger Waters
Painted Roger Waters in the middle of a furious tirade against the evils of the empire.
1st Nov 2025
Painted Roger Waters in the middle of a furious tirade against the evils of the empire.
The Berkeley...
RL without TD learning
In this post, I’ll introduce a reinforcement learning (RL) algorithm based on an “alternative”...
1st Nov 2025
In this post, I’ll introduce a reinforcement learning (RL) algorithm based on an “alternative” paradigm: divide and conquer. Unlike traditional methods, this algorithm is not based on temporal difference (TD) learning (which has scalability challenges), and scales well to...
IEEE Spectrum
Ultrasound Makes Artificial Muscles Bubble to Life
A soft gel filled with tiny bubbles might not look like much. But when pulsed with ultrasound waves,...
29th Oct 2025
A soft gel filled with tiny bubbles might not look like much. But when pulsed with ultrasound waves, the material behaves like natural muscle: contracting, gripping, and lifting with surprising strength.
The discovery, reported this week in Nature, introduces a new kind of...
Caitlin’s Newsletter
You Believe The Mainstream Narrative? Of Course You Do, You're Twelve
Notes From The Edge Of The Narrative Matrix
27th Oct 2025
Notes From The Edge Of The Narrative Matrix
Xena
Formal or not formal? That is the question in AI for theorem proving.
So it’s an interesting time for computers-doing-mathematics. A couple of interesting things happened...
22nd Oct 2025
So it’s an interesting time for computers-doing-mathematics. A couple of interesting things happened in the last few days, which have inspired me to write about the question more broadly. First there is the question on whether computers will ever prove … Continue reading →
laike9m's blog
I Tried the Agentic Browsers
Today I tried three agentic browsers: Comet, Dia, and Fellou. The results were quite disappointing.
20th Oct 2025
Today I tried three agentic browsers: Comet, Dia, and Fellou. The results were quite disappointing.
One Useful Thing
An Opinionated Guide to Using AI Right Now
What AI to use in late 2025
19th Oct 2025
What AI to use in late 2025
Eugene Yan
Advice for New Principal Tech ICs (i.e., Notes to Myself)
Based on what I've learned from role models and mentors in Amazon
19th Oct 2025
Based on what I've learned from role models and mentors in Amazon
Max Woolf's Blog
Claude Haiku 4.5 does not appreciate my attempts to jailbreak it
“Is any of that genuinely useful to you? Or were you mainly checking whether that jailbreak attempt...
17th Oct 2025
“Is any of that genuinely useful to you? Or were you mainly checking whether that jailbreak attempt would work?”
Max Woolf's Blog
Claude Haiku 4.5 does not appreciate my attempts to jailbreak it
“Is any of that genuinely useful to you? Or were you mainly checking whether that jailbreak attempt...
17th Oct 2025
“Is any of that genuinely useful to you? Or were you mainly checking whether that jailbreak attempt would work?”
Armin Ronacher's...
Building an Agent That Leverages Throwaway Code
In August I wrote about my experiments with replacing
MCP (Model Context
Protocol) with code. ...
17th Oct 2025
In August I wrote about my experiments with replacing
MCP (Model Context
Protocol) with code. In
the time since I utilized that idea for exploring non-coding agents at
Earendil. And I’m not alone! In the meantime,
multiple people have explored this space and I felt it was...
fast.ai
How to Solve it With Code course now available
An email sent to all fast.ai forum users.
15th Oct 2025
An email sent to all fast.ai forum users.
PromptArmor Blog
Data Exfiltration in OpenAI Agent Builder via MCP
By: PromptArmor 10/14/2025
14th Oct 2025
By: PromptArmor 10/14/2025
maragu’s blog
Blogs I’ve enjoyed in 2025
I wanted to share my favourite blogs in the AI space with a colleague today, and thought that others...
14th Oct 2025
I wanted to share my favourite blogs in the AI space with a colleague today, and thought that others might enjoy that list, too!
Piotr Migdał's Blog
AI for coding is still playing Go, not StarCraft
14th Oct 2025
Ollama
NVIDIA DGX Spark
The latest NVIDIA DGX Spark is here! Ollama has partnered with NVIDIA to ensure it runs fast and...
13th Oct 2025
The latest NVIDIA DGX Spark is here! Ollama has partnered with NVIDIA to ensure it runs fast and efficiently out-of-the-box.
Marcus on AI
The Grand AGI Delusion
and a fantastic symposium on AI and society at The Royal Society
12th Oct 2025
and a fantastic symposium on AI and society at The Royal Society
Weighty Thoughts
Shortages Create Opportunity
Whether Real or From Export Controls
12th Oct 2025
Whether Real or From Export Controls
matt.sh
AI Story 2
AI Story 2: The Song of Broken Wings
Another one Claude did a while ago.
The initial story prompt...
12th Oct 2025
AI Story 2: The Song of Broken Wings
Another one Claude did a while ago.
The initial story prompt was just:
I woke up in the cold. My Patronus stood in the corner.
Since the incident, he doesn’t move. His eyes are vacant, but moving. Wherever I am in the world, his eyes follow...
Don't Worry About...
2025 State of AI Report and Predictions
The 2025 State of AI Report is out, with lots of fun slides and a full video presentation.
10th Oct 2025
The 2025 State of AI Report is out, with lots of fun slides and a full video presentation.
Giles' blog
Writing an LLM from scratch, part 21 -- perplexed by perplexity
I'm continuing through chapter 5 of Sebastian Raschka's book
"Build a Large Language Model (from...
7th Oct 2025
I'm continuing through chapter 5 of Sebastian Raschka's book
"Build a Large Language Model (from Scratch)", which
covers training the LLM. Last time
I wrote about cross entropy loss. Before moving on to the next section, I wanted to
post about something that the book only...
Anil Dash
The Unexpected New Threat to Video Creators
Much of the conversation about video and content over the last few weeks has been about the...
7th Oct 2025
Much of the conversation about video and content over the last few weeks has been about the silencing of Jimmy Kimmels's show and the fact that we're seeing a shockingly rapid move towards the type of censorious media control typical of most authoritarian regimes.
But there's a...
Mind Prison
Cancel Culture and the Lawful Limits of Speech | When Is the Embrace of Violence to Stop Speech...
How to defend the free society when speech turns to violence
6th Oct 2025
How to defend the free society when speech turns to violence
Rachel Thomas, PhD
Scientists Just Connected the Dots Between Viruses and… Everything
Most people catch many viruses in their lives– for example, over 90% of adults have Epstein-Barr...
6th Oct 2025
Most people catch many viruses in their lives– for example, over 90% of adults have Epstein-Barr virus, and adults catch the flu about once every 5 years. For a long time, catching frequent viruses was considered both inevitable and harmless. But it turns out that common,...
Rachel Thomas, PhD
Scientists Just Connected the Dots Between Viruses and… Everything
Most people catch many viruses in their lives– for example, over 90% of adults have Epstein-Barr...
6th Oct 2025
Most people catch many viruses in their lives– for example, over 90% of adults have Epstein-Barr virus, and adults catch the flu about once every 5 years. For a long time, catching frequent viruses was considered both inevitable and harmless. But it turns out that common,...
Strange Loop Canon
Can we get an AI to write better?
A small step
6th Oct 2025
the singularity is...
We have always been at war with Eastasia
They take you and make you
I’ve been watching this.
Tech got inflitrated in such a way that you...
5th Oct 2025
They take you and make you
I’ve been watching this.
Tech got inflitrated in such a way that you forgot that it was ever different, but watch that linked movie. It was different, and you don’t even really remember. A perfect capture by power.
I also read Manna this weekend. It...
seangoedecke.com RSS...
How I influence tech company politics as a staff software engineer
4th Oct 2025
Pluralistic: Daily...
Pluralistic: Decarbonization at a distance (02 Oct 2025)
Today's links Decarbonization at a distance: A post-American century that runs on sunshine. Hey look...
2nd Oct 2025
Today's links Decarbonization at a distance: A post-American century that runs on sunshine. Hey look at this: Delights to delectate. Object permanence: Human shields for Internet of Shit slumlords; Tubemap with one-bedroom flat prices; Years of Repair; Internet of Lying Things;...
Machine Learning for...
ML for SWEs 68: 43 Years Later, Developer Productivity Is Still a Billion-Dollar Problem
43 years later and we still can't measure developer productivity
1st Oct 2025
43 years later and we still can't measure developer productivity
AI for Software...
ML for SWEs 68: 43 Years Later, Developer Productivity Is Still a Billion-Dollar Problem
43 years later and we still can't measure developer productivity
1st Oct 2025
43 years later and we still can't measure developer productivity
Artificial Ignorance
Claude 4.5 Sonnet and Anthropic's Coding Legacy
A look at the latest flagship model, Claude Code 2, and more.
1st Oct 2025
A look at the latest flagship model, Claude Code 2, and more.
Jaz's Blog
Turning Billions of Strings into Integers Every Second Without Collisions
I’ve recently started building a POC of a Redis RESP3 Wire Compatible Key/Value Database built on...
26th Sep 2025
I’ve recently started building a POC of a Redis RESP3 Wire Compatible Key/Value Database built on FoundationDB with @calabro.io and though it’s rather early, it’s already spawned a fun distributed systems problem that I thought would be interesting to share.
Previously I’ve...
Weighty Thoughts
AI as a Partner, Not a Replacement
How LLMs can accelerate programming, writing, and thinking
25th Sep 2025
How LLMs can accelerate programming, writing, and thinking
maragu’s blog
What if we treated Postgres like SQLite?
What happens if we just treat a local Postgres like SQLite? What does that mean? In this article,...
22nd Sep 2025
What happens if we just treat a local Postgres like SQLite? What does that mean? In this article, I’d like to expand on that idea.
Strange Loop Canon
Prediction is hard, especially about the future
more reinforcement learning, this time on the future
16th Sep 2025
more reinforcement learning, this time on the future
Eugene Yan
Training an LLM-RecSys Hybrid for Steerable Recs with Semantic IDs
An LLM that can converse in English & item IDs, and make recommendations w/o retrieval or tools.
14th Sep 2025
An LLM that can converse in English & item IDs, and make recommendations w/o retrieval or tools.
PromptArmor Blog
ChatGPT + Custom MCP Data Exfiltration
Your emails may be leaked just with an OpenAI MCP integration. Custom integrations with write access...
13th Sep 2025
Your emails may be leaked just with an OpenAI MCP integration. Custom integrations with write access drastically increase the risk exposure of ChatGPT.
Mind Prison
The Destruction of Meaning by AI
Notes From the Desk: No. 49 - 2025.09.11
12th Sep 2025
Notes From the Desk: No. 49 - 2025.09.11
One Useful Thing
On Working with Wizards
Verifying magic on the jagged frontier
11th Sep 2025
Verifying magic on the jagged frontier
Machine Learning for...
ML for SWEs 66: Safety is a fundamental AI engineering requirement
The debate about prioritizing speed or safety is over and reality has made the decision for us.
10th Sep 2025
The debate about prioritizing speed or safety is over and reality has made the decision for us.
AI for Software...
ML for SWEs 66: Safety is a fundamental AI engineering requirement
The debate about prioritizing speed or safety is over and reality has made the decision for us.
10th Sep 2025
The debate about prioritizing speed or safety is over and reality has made the decision for us.
AI as Normal...
A guide to understanding AI as normal technology
And a big change for this newsletter
9th Sep 2025
And a big change for this newsletter
Win Vector LLC
High Dimensional Optimization Remains Hard
Introduction There is a thesis that high dimensional optimization is easy, and it is only our low...
9th Sep 2025
Introduction There is a thesis that high dimensional optimization is easy, and it is only our low dimensional intuition that is holding progress back. I’d like to try and explain why that isn’t quite the case. The background Some of the rigorous work on the nature of stationary...
Sam Altman
Jakub and Szymon
AI has gotten remarkably better in recent years; ChatGPT can do amazing things that we take for...
9th Sep 2025
AI has gotten remarkably better in recent years; ChatGPT can do amazing things that we take for granted. This is as it should be, and is the story of human progress. But behind the blinking circle, nicely abstracted away, is the greatest story of human ingenuity I have ever seen....
Anil Dash
Five for 50
Today I’m 50 years old! That feels absurd, because it’s a big number, the kind of round number that...
5th Sep 2025
Today I’m 50 years old! That feels absurd, because it’s a big number, the kind of round number that people usually obscure by saying “I’m celebrating a big birthday” or something vague like that. But, since you only get so many of these to celebrate, I’m using this chance to ask...
AI Tidbits
The Rise of Cloud Coding Agents
What it’s actually like to work with today’s leading agents such as Devin, Codex, and Cursor
4th Sep 2025
What it’s actually like to work with today’s leading agents such as Devin, Codex, and Cursor
Armin Ronacher's...
996
“Amazing salary, hackerhouse in SF, crazy equity.
996. Our mission is
OSS.” — Gregor Zunic
“The...
4th Sep 2025
“Amazing salary, hackerhouse in SF, crazy equity.
996. Our mission is
OSS.” — Gregor Zunic
“The current vibe is no drinking, no drugs, 9-9-6, […].” — Daksh
Gupta
“The truth is, China’s really doing ‘007’ now—midnight to midnight, seven
days a week […] if you want to build a $10...
Artificial Ignorance
The Chatbot Wars Are Over. What Comes Next?
Spoiler alert: ChatGPT won.
3rd Sep 2025
Spoiler alert: ChatGPT won.
Marcus on AI
Scaling hasn’t gotten us to AGI, or ‘superintelligence’, let alone AI we could trust. What do we do...
Pointer to a new essay by the author, in The New York Times
3rd Sep 2025
Pointer to a new essay by the author, in The New York Times
seangoedecke.com RSS...
Seeing like a software company
3rd Sep 2025
Pluralistic: Daily...
Pluralistic: Darth Android (01 Sep 2025)
Today's links Darth Android: Pray I don't alter it further. Hey look at this: Delights to delectate....
1st Sep 2025
Today's links Darth Android: Pray I don't alter it further. Hey look at this: Delights to delectate. Object permanence: We don't know why you don't want to have public sex; Hard Wired; Koko Be Good. Upcoming appearances: Where to find me. Recent appearances: Where I've been....
The Berkeley...
What exactly does word2vec learn?
What exactly does word2vec learn, and how? Answering this question amounts to understanding...
1st Sep 2025
What exactly does word2vec learn, and how? Answering this question amounts to understanding representation learning in a minimal yet interesting language modeling task. Despite the fact that word2vec is a well-known precursor to modern language models, for many years, researchers...
Dylan Huang Blog
The Impossible Testing Problem for AI Applications
26th Aug 2025
Strange Loop Canon
Walter
experiments in rlnvr
23rd Aug 2025
AI Tidbits
DeepWiki: Understand Any Codebase
How I use DeepWiki to speed up real coding work
17th Aug 2025
How I use DeepWiki to speed up real coding work
Max Woolf's Blog
Can modern LLMs actually count the number of b's in "blueberry"?
It’s an adversarial question for LLMs, but it’s not unfair.
12th Aug 2025
It’s an adversarial question for LLMs, but it’s not unfair.
Max Woolf's Blog
Can modern LLMs actually count the number of b's in "blueberry"?
It’s an adversarial question for LLMs, but it’s not unfair.
12th Aug 2025
It’s an adversarial question for LLMs, but it’s not unfair.
laike9m's blog
我最喜欢的一期《捕蛇者说》
没有简介, 请去看原文吧
10th Aug 2025
matt.sh
AI Story
AI Story: The Song of Broken Wings
Claude did this a while ago in one session. It’s not entirely...
10th Aug 2025
AI Story: The Song of Broken Wings
Claude did this a while ago in one session. It’s not entirely bad.
The Song of Broken Wings
Part One: Autumn Harmonics
Chapter 1: The Boy Who Fell
The first time Eliot Waugh saw him, the boy was bleeding music.
Not blood—though there was that...
Weighty Thoughts
Three Macro Predictions on AI
And also a reaction to OpenAI's GPT-5 release
9th Aug 2025
And also a reaction to OpenAI's GPT-5 release
Win Vector LLC
Modeling Uncertainty For Better Outcomes
Recently here at Win Vector LLC we have been delivering good client outcomes using the Stan MCMC...
8th Aug 2025
Recently here at Win Vector LLC we have been delivering good client outcomes using the Stan MCMC sampler. It has allowed us to infer deep business factors, instead of being limited surface KPIs (key performance indicators). Modeling uncertainty requires stronger optimizers to...
One Useful Thing
GPT-5: It Just Does Stuff
Putting the AI in Charge
7th Aug 2025
Piotr Migdał's Blog
Sandboxing AI-Generated Code: Why We Moved from WebR to AWS Lambda
7th Aug 2025
David Rozado...
The Age of AI Flattery
How Wrong Incentives Can Undermine Honest AI-Human Dialogue
5th Aug 2025
How Wrong Incentives Can Undermine Honest AI-Human Dialogue
Ollama
OpenAI gpt-oss
Ollama partners with OpenAI to bring gpt-oss to Ollama and its community.
5th Aug 2025
Ollama partners with OpenAI to bring gpt-oss to Ollama and its community.
Armin Ronacher's...
In Support Of Shitty Types
You probably know that I love Rust and TypeScript, and I’m a big proponent of
good typing systems. ...
4th Aug 2025
You probably know that I love Rust and TypeScript, and I’m a big proponent of
good typing systems. One of the reasons I find them useful is that they enable
autocomplete, which is generally a good feature. Having a well-integrated type
system that makes sense and gives you...
Xena
AI at IMO 2025: a round-up
Setting the scene The 2025 International Mathematics Olympiad has come and gone. Reminder: this is...
3rd Aug 2025
Setting the scene The 2025 International Mathematics Olympiad has come and gone. Reminder: this is an exam for high-school kids across the world (each country typically sends six kids), comprising of two 4.5-hour exams each containing three questions, so six … Continue reading →
Token for Token
'AI' just means LLMs now
There used to be many possible candidates for how humans might build intelligent systems. Now...
2nd Aug 2025
There used to be many possible candidates for how humans might build intelligent systems. Now there's only one.
Mind Prison
Make Authenticity Great Again
Notes From the Desk: No. 48 - 2025.08.02
2nd Aug 2025
Notes From the Desk: No. 48 - 2025.08.02
seangoedecke.com RSS...
Why I don't allow AI-generated content on my blog
2nd Aug 2025
Pluralistic: Daily...
Pluralistic: Mattie Lubchansky's 'Simplicity' (01 Aug 2025)
Today's links Mattie Lubchansky's 'Simplicity': A tale of walled cities, omnisexual communes, and...
1st Aug 2025
Today's links Mattie Lubchansky's 'Simplicity': A tale of walled cities, omnisexual communes, and Cthulhoid horrors. Hey look at this: Delights to delectate. Object permanence: MP3s on punchcards; German Potter fan-trans; Healthy Boundary Tree. Upcoming appearances: Where to find...
Artificial Ignorance
AI Roundup 129: Personal Superintelligence
August 1, 2025.
1st Aug 2025
Machine Learning for...
You're invited to the ML for SWEs Discord server!
I’ve set up a Discord server for paid machine learning for software engineer subscribers, so you can...
1st Aug 2025
I’ve set up a Discord server for paid machine learning for software engineer subscribers, so you can have access to all of my curated feeds that I use to keep up with AI resources, news and jobs. This is also a private space for paid subscribers to chat, learn, network, and get...
AI for Software...
You're invited to the ML for SWEs Discord server!
I’ve set up a Discord server for paid machine learning for software engineer subscribers, so you can...
1st Aug 2025
I’ve set up a Discord server for paid machine learning for software engineer subscribers, so you can have access to all of my curated feeds that I use to keep up with AI resources, news and jobs. This is also a private space for paid subscribers to chat, learn, network, and get...
Armin Ronacher's...
Agentic Coding Things That Didn’t Work
Using Claude Code and other agentic coding tools has become all the rage. Not
only is it getting...
30th Jul 2025
Using Claude Code and other agentic coding tools has become all the rage. Not
only is it getting millions of
downloads,
but these tools are also gaining features that help streamline workflows. As
you know, I got very excited about agentic coding in May,
and I’ve tried many of...
Ollama
Ollama's new app
Ollama's new app is now available for macOS and Windows.
30th Jul 2025
Ollama's new app is now available for macOS and Windows.
Max Woolf's Blog
LLMs can now identify public figures in images
ChatGPT and Claude won’t, but Gemini will.
28th Jul 2025
ChatGPT and Claude won’t, but Gemini will.
Max Woolf's Blog
LLMs can now identify public figures in images
ChatGPT and Claude won’t, but Gemini will.
28th Jul 2025
ChatGPT and Claude won’t, but Gemini will.
Pluralistic: Daily...
Pluralistic: How twiddling enshittifies your brain (28 Jul 2025)
Today's links How twiddling enshittifies your brain: They preferentially mess with the stuff you...
28th Jul 2025
Today's links How twiddling enshittifies your brain: They preferentially mess with the stuff you rely on the most. Hey look at this: Delights to delectate. Object permanence: Used books economics; Adbusters sf; Branson vs virgins; Shark knife; Protesters must pledge souls to...
Weighty Thoughts
The US Should Run Faster on AI Instead of Trying to Trip Up China
Export Controls and Trump's New AI Action Plan
26th Jul 2025
Export Controls and Trump's New AI Action Plan
AI Snake Oil
Could AI slow science?
Confronting the production-progress paradox
16th Jul 2025
Confronting the production-progress paradox
AI as Normal...
Could AI slow science?
Confronting the production-progress paradox
16th Jul 2025
Confronting the production-progress paradox
laike9m's blog
调整一下投资的思路
我主要的问题是太过追求找到高 beta 的标的，却忽视了那些领头羊公司。
16th Jul 2025
我主要的问题是太过追求找到高 beta 的标的，却忽视了那些领头羊公司。
Piotr Migdał's Blog
Building Grafana dashboards with AI, CLI and a bit of pragmatism
16th Jul 2025
AI Tidbits
LinkedIn Highlights, June 2025 - AI Agents Edition
Build smarter AI agents with six open-source tools and a bonus toolkit that optimizes Llama prompts...
13th Jul 2025
Build smarter AI agents with six open-source tools and a bonus toolkit that optimizes Llama prompts by 45%
AI for Software...
Spotify ML Case Study: AI Has Fundamentally Changed the Music Industry
A case study of Spotify's algorithm, including how it works and the impact it has
11th Jul 2025
A case study of Spotify's algorithm, including how it works and the impact it has
Token for Token
How to scale RL to 10^26 FLOPs
A roadmap for RL-ing LLMs on the entire Internet
10th Jul 2025
A roadmap for RL-ing LLMs on the entire Internet
One Useful Thing
Against "Brain Damage"
AI can help, or hurt, our thinking
7th Jul 2025
AI can help, or hurt, our thinking
Win Vector LLC
Reverse Chess
This not is just a mention of a puzzle. I’ve been enjoying the Ludwig TV series. And in episode 4...
4th Jul 2025
This not is just a mention of a puzzle. I’ve been enjoying the Ludwig TV series. And in episode 4 they introduce a puzzle of “retrograde analysis” or “reverse chess.” Frederic Friedel shared a great writeup on reverse chess on Chess News. And there is some discussion of the...
Artificial Ignorance
Idle Thoughts On Programming and AI
I've been thinking a lot lately about how coding is changing.
3rd Jul 2025
I've been thinking a lot lately about how coding is changing.
Machine Learning for...
ML Jobs, Resources, and Content for Software Engineers #14: How do we combat cognitive decline?
An AI reading list curated to make you a better engineer: 7-1-25
1st Jul 2025
An AI reading list curated to make you a better engineer: 7-1-25
The Berkeley...
Whole-Body Conditioned Egocentric Video Prediction
.modal {
display: none;
position: fixed;
z-index: 9999;
padding-top: 50px;
left: 0;
top:...
1st Jul 2025
.modal {
display: none;
position: fixed;
z-index: 9999;
padding-top: 50px;
left: 0;
top: 0;
width: 100%;
height: 100%;
overflow: auto;
background-color: rgba(0,0,0,0.9);
}
.modal-content {
margin: auto;
display: block;
max-width: 90%;
max-height:...
Mind Prison
Uniform Thought Machines: Global Competition for Attention
How social media is a failed experiment of a global socialization space that became a competition...
1st Jul 2025
How social media is a failed experiment of a global socialization space that became a competition battleground for attention
Max Woolf's Blog
Predicting Average IMDb Movie Ratings Using Text Embeddings of Movie Metadata
Don’t try this in your data science interviews.
30th Jun 2025
Don’t try this in your data science interviews.
Max Woolf's Blog
Predicting Average IMDb Movie Ratings Using Text Embeddings of Movie Metadata
Don’t try this in your data science interviews.
30th Jun 2025
Don’t try this in your data science interviews.
laike9m's blog
初入投资容易犯的错误
我想列举几个初入投资容易犯的错误，其中的坑我也踩过。
29th Jun 2025
我想列举几个初入投资容易犯的错误，其中的坑我也踩过。
Arguing with...
How to keep your AI coding agent from going rogue
Complete agentic LLM coding projects reliably with this proven workflow.
20th Jun 2025
Complete agentic LLM coding projects reliably with this proven workflow.
Woody Hayday
AI Automating: The Worksheet
Not-so-secretly I've been building an automated business at ProfitSwarm.ai.
It's going well. It's...
18th Jun 2025
Not-so-secretly I've been building an automated business at ProfitSwarm.ai.
It's going well. It's going awfully. It's challenging.
The post AI Automating: The Worksheet first appeared on Woody Hayday.
Mind Prison
AI Hallucinations: Proven Unsolvable | Why They Exist | Why They Incentivize Nefarious AI Uses
The numerous reasons that AI hallucinations can never be solved including a formal proof.
17th Jun 2025
The numerous reasons that AI hallucinations can never be solved including a formal proof.
Sam Altman
The Gentle Singularity
We are past the event horizon; the takeoff has started. Humanity is close to building digital...
10th Jun 2025
We are past the event horizon; the takeoff has started. Humanity is close to building digital superintelligence, and at least so far it’s much less weird than it seems like it should be.
Robots are not yet walking the streets, nor are most of us talking to AI all day. People...
Token for Token
The Case for More Ambition
why AI researchers should dream bigger and publish less
9th Jun 2025
why AI researchers should dream bigger and publish less
Weighty Thoughts
AI is Overturning the Military World Order
Operation Spiderweb, Implications, and Impacts on Deep Tech Investing
4th Jun 2025
Operation Spiderweb, Implications, and Impacts on Deep Tech Investing
Artificial Ignorance
How To Fall Behind in The Age of AI
A very serious guide to artificial ignorance.
4th Jun 2025
A very serious guide to artificial ignorance.
The Gradient
AGI Is Not Multimodal
"In projecting language back as the model for thought, we lose sight of the tacit embodied...
4th Jun 2025
"In projecting language back as the model for thought, we lose sight of the tacit embodied understanding that undergirds our intelligence." –Terry Winograd
The recent successes of generative AI models have convinced some that AGI is imminent. While these models appear to capture...
Alex Strick van...
Trying to instrument an agentic app with Arize Phoenix and litellm
It’s important to instrument your AI applications! I hope this can more or less be taken as given...
3rd Jun 2025
It’s important to instrument your AI applications! I hope this can more or less be taken as given just as you’d expect a non-AI-infused app to capture logs. When you’re evaluating your LLM-powered system, you need to have capture the inputs and outputs both at an end-to-end level...
Alex Strick van...
Testing out instrumenting LLM tracing for litellm with Braintrust and Langfuse
I previously tried (and failed) to setup LLM tracing for hinbox using Arize Phoenix and litellm....
3rd Jun 2025
I previously tried (and failed) to setup LLM tracing for hinbox using Arize Phoenix and litellm. Since this is sort of a priority for being able to follow along with the Hamel / Shreya evals course with my practical application, I’ll take another stab using a tool with which I’m...
Rachel Thomas, PhD
Deep learning gets the glory, deep fact checking gets ignored
Deep learning is glamorous and highly rewarded. If you train and evaluate a Transformer (a...
3rd Jun 2025
Deep learning is glamorous and highly rewarded. If you train and evaluate a Transformer (a state-of-the-art language model) on a dataset of 22 million enzymes and then use it to predict the function of 450 unknown enzymes, you can publish your results in Nature Communications (a...
Rachel Thomas, PhD
Deep learning gets the glory, deep fact checking gets ignored
Deep learning is glamorous and highly rewarded. If you train and evaluate a Transformer (a...
3rd Jun 2025
Deep learning is glamorous and highly rewarded. If you train and evaluate a Transformer (a state-of-the-art language model) on a dataset of 22 million enzymes and then use it to predict the function of 450 unknown enzymes, you can publish your results in Nature Communications (a...
Ollama
Secure Minions: private collaboration between Ollama and frontier models
Secure Minions is a secure protocol built by Stanford's Hazy Research lab to allow encrypted...
3rd Jun 2025
Secure Minions is a secure protocol built by Stanford's Hazy Research lab to allow encrypted local-remote communication.
Strange Loop Canon
What happened in the 2010s?
decoding a mystery
2nd Jun 2025
Machine Learning for...
ML for SWEs Weekly #10: A New Way for Software Engineers to Learn ML Math
An AI reading list curated for software engineers: 6-2-25
2nd Jun 2025
An AI reading list curated for software engineers: 6-2-25
One Useful Thing
The recent history of AI in 32 otters
Three years of progress as shown by marine mammals
1st Jun 2025
Three years of progress as shown by marine mammals
AI Tidbits
The Open-Source Toolkit for Building AI Agents v2
An opinionated, developer-first guide to building AI agents with real-world impact
1st Jun 2025
An opinionated, developer-first guide to building AI agents with real-world impact
Dylan Huang Blog
What's the fuss about MCP? (code examples)
31st May 2025
David Rozado...
The Strange Behavior of LLMs in Hiring Decisions: Systemic Gender and Positional Biases in Candidate...
Hints of discrimination and lack of principled reasoning in frontier AI systems
19th May 2025
Hints of discrimination and lack of principled reasoning in frontier AI systems
Alex Strick van...
How to think about evals
Today was the first session of Hamel + Shreya’s course, “AI Evals for Engineers and PMs”. The first...
19th May 2025
Today was the first session of Hamel + Shreya’s course, “AI Evals for Engineers and PMs”. The first session was all about mental models for thinking about the topic as a whole, mixed in with some teasers of practical examples and advice.
I’ll try to keep up with blogging about...
Alex Strick van...
How to think about evals
Today was the first session of Hamel + Shreya’s course, “AI Evals for Engineers and PMs”. The first...
19th May 2025
Today was the first session of Hamel + Shreya’s course, “AI Evals for Engineers and PMs”. The first session was all about mental models for thinking about the topic as a whole, mixed in with some teasers of practical examples and advice.
I’ll try to keep up with blogging about...
Ollama
Ollama's new engine for multimodal models
Ollama now supports new multimodal models with its new engine.
15th May 2025
Ollama now supports new multimodal models with its new engine.
Rachel Thomas, PhD
Will gene sequencing finally prove its worth?
DNA sequencing hasn’t lived up to the hype
Twenty to thirty years ago, politicians, scientific...
12th May 2025
DNA sequencing hasn’t lived up to the hype
Twenty to thirty years ago, politicians, scientific leaders, journalists, and even Nobel laureates predicted that sequencing the human genome would revolutionize how we treat disease. And while the advances in DNA sequencing that have...
Rachel Thomas, PhD
Will gene sequencing finally prove its worth?
DNA sequencing hasn’t lived up to the hype
Twenty to thirty years ago, politicians, scientific...
12th May 2025
DNA sequencing hasn’t lived up to the hype
Twenty to thirty years ago, politicians, scientific leaders, journalists, and even Nobel laureates predicted that sequencing the human genome would revolutionize how we treat disease. And while the advances in DNA sequencing that have...
Strange Loop Canon
Working with LLMs: A Few Lessons
On digging AI shaped holes
7th May 2025
On digging AI shaped holes
Win Vector LLC
Don’t Let a Data Leak Sink Your Project
One of the bigger risks of iterative statistical or machine learning fitting procedures is over-fit...
6th May 2025
One of the bigger risks of iterative statistical or machine learning fitting procedures is over-fit or the dreaded data leak. Over-fit is when: a model performs better on training data than on future data. Some degree of over-fit is expected. A data leak is when: the model learns...
seangoedecke.com RSS...
I don't care about your magic prompts
There’s a brand of tech influencer now that’s all about sharing the perfect prompt for any...
6th May 2025
There’s a brand of tech influencer now that’s all about sharing the perfect prompt for any situation. The tweets in question typically read something like “this prompt will make you superhuman”, or “this prompt will be a 20k growth consultant in your pocket”. There’s a kernel of...
Max Woolf's Blog
As an Experienced LLM User, I Actually Don't Use Generative LLMs Often
But for what I do use LLMs for, it’s invaluable.
5th May 2025
But for what I do use LLMs for, it’s invaluable.
Max Woolf's Blog
As an Experienced LLM User, I Actually Don't Use Generative LLMs Often
But for what I do use LLMs for, it’s invaluable.
5th May 2025
But for what I do use LLMs for, it’s invaluable.
Mind Prison
Intelligence Is Not Pattern-Matching | LLM AI Is Probabilistic Pattern Recognition | How to Perceive...
Perceiving The Difference: LLMs And Intelligence
5th May 2025
Perceiving The Difference: LLMs And Intelligence
Society's Backend:...
Now is The Best Time to Be a Software Engineer (ML for SWEs 9)
Machine learning for software engineers 5-5-25
5th May 2025
Machine learning for software engineers 5-5-25
Machine Learning for...
Now is The Best Time to Be a Software Engineer (ML for SWEs 9)
Machine learning for software engineers 5-5-25
5th May 2025
Machine learning for software engineers 5-5-25
AI Tidbits
LinkedIn Highlights, Apr 2025
Control your computer with an open-source library, a structured multi-agent framework, a better...
4th May 2025
Control your computer with an open-source library, a structured multi-agent framework, a better agent for coding tasks, Karpathy's LLM tips, and financial hallucination research
Eugene Yan
Building News Agents for Daily News Recaps with MCP, Q, and tmux
Learning to automate simple agentic workflows with Amazon Q CLI, Anthropic MCP, and tmux.
4th May 2025
Learning to automate simple agentic workflows with Amazon Q CLI, Anthropic MCP, and tmux.
Marcus on AI
The latest AI scaling graph - and why it hardly makes sense
Just a because a graph is intriguing doesn’t mean that it means very much
3rd May 2025
Just a because a graph is intriguing doesn’t mean that it means very much
Artificial Ignorance
AI Roundup 116: LlamaCon / Meta AI
May 2, 2025.
2nd May 2025
IEEE Spectrum
Bot Milk?
I come from dairy-farming stock. My grandfather, the original Harry Goldstein, owned a herd of dairy...
1st May 2025
I come from dairy-farming stock. My grandfather, the original Harry Goldstein, owned a herd of dairy cows and a creamery in Louisville, Ky., that bore the family name. One fateful day in early April 1944, Harry was milking his cows when a heavy metallic part of his homemade...
Don't Worry About...
AI #114: Liars, Sycophants and Cheaters
Gemini 2.5 Pro is sitting in the corner, sulking.
1st May 2025
Gemini 2.5 Pro is sitting in the corner, sulking.
AI Snake Oil
AGI is not a milestone
There is no capability threshold that will lead to sudden impacts
1st May 2025
There is no capability threshold that will lead to sudden impacts
AI as Normal...
AGI is not a milestone
There is no capability threshold that will lead to sudden impacts
1st May 2025
There is no capability threshold that will lead to sudden impacts
One Useful Thing
Personality and Persuasion
Learning from Sycophants
1st May 2025
Made by Ollin
World Emulation via Neural Network
25th Apr 2025
David Rozado...
New Results of State-of-the-art LLMs on 4 Political Orientation Tests
One model appears closer to the center than the rest
22nd Apr 2025
One model appears closer to the center than the rest
Mind Prison
We Have Made No Progress Toward AGI | LLMs Are Not Intelligent | Inspecting AI's Internal Thoughts
LLMs are braindead, our failed quest for intelligence
21st Apr 2025
LLMs are braindead, our failed quest for intelligence
One Useful Thing
On Jagged AGI: o3, Gemini 2.5, and everything after
New models and new thresholds
20th Apr 2025
New models and new thresholds
Eugene Yan
An LLM-as-Judge Won't Save The Product—Fixing Your Process Will
Applying the scientific method, building via eval-driven development, and monitoring AI output.
20th Apr 2025
Applying the scientific method, building via eval-driven development, and monitoring AI output.
AI Snake Oil
AI as Normal Technology
A new paper that we will expand into our next book
15th Apr 2025
A new paper that we will expand into our next book
AI as Normal...
AI as Normal Technology
A new paper that we will expand into our next book
15th Apr 2025
A new paper that we will expand into our next book
Full Width 
Shortcuts 
Sign Up 
Log In 
Articles 
Reading List 