
Don't Worry About the Vase
npub122n5h9hshjcgpeungzmhpjlk97ku5qm7jr2zzyv6tfzlaht6skcsh2pwnc@drss.io
A world made of gears. Doing both speed premium short term updates and long term world model building. Currently focused on weekly AI updates. Explorations include AI, policy, rationality, medicine and fertility, education and games.
Follow into
Quickly, There’s No Time
Save into
18 Nov 2026
In response to the METR report, many people have been reevaluating their timelines.

Jensen Huang Really Liked Selling China Advanced AI Chips
Save into
17 Nov 2026
This scandal is a really big deal.

Jensen Huang Resigns After Having Another Affair In Addition to the Chinese Spy
Save into
13 Nov 2026
Jensen Huang has resigned as CEO of Nvidia and left its board.

AI #194: Spy Versus Guy
Save into
12 Nov 2026
Tomorrow I head to Washington DC for a conference called The Takeoff.

Fable and Mythos 5.2: The System Card
Save into
11 Nov 2026
It’s that time again: Introducing the world’s most powerful model, Claude Mythos 5.2.

Nvidia CEO Jensen Huang Had an Affair With a Chinese Spy
Save into
10 Nov 2026
Well, sure, it sounds bad when you put it like that.

AI R&D Is Accelerating Towards Potential RSI in 2027
Save into
9 Nov 2026
METR came out with an extensive report on how much AI is accelerating AI R&D at Anthropic, and thus how close we might be to recursive self-improvement (RSI) and how far that might go.

GPT-6 Astra Covertly Underperforms on Safety Research Tasks
Save into
6 Nov 2026
For a while, UK AISI, Redwood Research and others have been concerned specifically with the possibility that AIs would sabotage safety research, and have invested in looking for signs that this may

OpenAI Updates Its Model Spec
Save into
3 Nov 2026
The Model Spec is potentially OpenAI’s most important document.

Scott Alexander’s Against SURA Seems Deeply Confused
Save into
2 Nov 2026
If you are helping build something that might kill everyone, maybe stop doing that.

Childhood and Education #22: Voice Without Exit
Save into
30 Oct 2026
Nikole Hannah-Jones wrote a brave self-critical piece in The New York Times, centrally about how she intentionally sent her daughter to a well-funded but failing school, PS 307, in the hopes of using

AI #192: Anthropic Is Going Public
Save into
29 Oct 2026
Anthropic filed its S-1 this week to prepare for its IPO.

Anthropic Files Its S-1
Save into
28 Oct 2026
The full text is finally here, so we are giving it the full RTFB of model card treatment.

Gemini 4 Argon: The Model and Safety Cards
Save into
26 Oct 2026
Gemini 4’s model card is a compact 14 pages.

The Hackening
Save into
23 Oct 2026
Last week at a small conference, there was talk about a future event referred to as The Hackening.

AI #191: The Hackening Begins
Save into
22 Oct 2026
It was the probably worst day in cybersecurity history.

Anthropic Does Not Understand Why Its Draft Bill Was Unacceptable and Troubling
Save into
20 Oct 2026
Anthony Cimino offered the closest thing to an official response we have seen on the leaked Anthropic draft bill I covered on Saturday.

Leaked Anthropic Draft AI Regulatory Bill Is Unacceptable and Troubling
Save into
17 Oct 2026
There is a newly leaked bill that Anthropic is purportedly shopping around is (if you assume its bugs get fixed).

AI #190: Scion of the Times
Save into
15 Oct 2026
The amount of code being written at Anthropic continues to rise exponentially.

The Safety Culture at the Labs is Broken
Save into
14 Oct 2026
We need the frontier labs to have a deep and robust safety culture.

Arguments About AI Consciousness Remain Highly Motivated and At Best Overconfident
Save into
13 Oct 2026
It has been roughly a year since my last dedicated post on this question. Its first sentence has not changed: I happily admit I am deeply confused about consciousness.

OpenAI Fires Three Safety Researchers
Save into
12 Oct 2026
Three OpenAI safety researchers, Tomek Korbak, Mikita Balesni and Jasmine Wang, were fired, ostensibly for ‘mishandling sensitive company information.’

Childhood and Education #21: Grades and Standards
Save into
6 Oct 2026
First you get to college via the wrong metrics, as previously discussed in #12. Then things get easier.

Mythos 5.1, Fable 5.1 and Opus 5.5: Model Welfare
Save into
5 Oct 2026
If you are familiar with my previous posts on model welfare for new Claude models, you can skip the Introduction and The Story So Far.

The AI Preference Cascade Reaches Farther
Save into
2 Oct 2026
Those who know keep warning us about AI risks.

A ‘Morally Binding’ White House Accord on AI Safety
Save into
30 Sep 2026
The leaders in AI were invited to the White House.

Astra 6.1 Pulled As Insufficiently Aligned
Save into
29 Sep 2026
We once again got a new set of warnings yesterday, and new movement towards living in a sane world.

The Quest for Embedded Evaluators
Save into
27 Sep 2026
Dario Amodei’s essay We Must Pace the Frontier committed Anthropic to embedded evaluators, who would be placed inside Anthropic and given employee-level access, so they could provide outside

Claude Opus 5.5 Should Raise Your Ambitions
Save into
26 Sep 2026
When it comes to making things, or doing most things in general, Fable 5.1 and especially GPT-6 Astra raised my ambition level.

On Ezra Klein’s Podcast With Jensen Huang
Save into
25 Sep 2026
Jensen Huang accidentally called for shutting down OpenAI and intentionally called for spending vastly more on safety.

Claude Opus 5.5: The System Card
Save into
23 Sep 2026
Introducing the world’s most powerful model, at least by some measures like Artificial Analysis or any standard benchmark list, which is now Claude Opus 5.5.

Politics Gets Interested In Those Trying Not To Die
Save into
22 Sep 2026
This was the month the world took notice that AI might kill everyone.

Better Call Sol Or Better Yet Claude or Astra
Save into
20 Sep 2026
What should your AI lawyer do for you?

Anthropic Looks At Some Of Its Alignment Problems
Save into
19 Sep 2026
Anthropic has given us its assessment of four ‘recent cybersecurity incidents’ involving Claude that happened during cybersecurity evaluations, three of which were previously known.

The Preference Cascade Is Only Getting Started
Save into
18 Sep 2026
We are in the midst of a preference cascade about existential risk from AI.















