James

[James @JayGeraghty]

Good morning. Off on hols ✈️

[a16z @a16z]

Texas builds. Its ~14 GW under construction, the most in America, is just ~15% of the state's private non-residential spending.

In New Mexico and Wyoming, data centers might as well be the construction industry: ~60% of spending, on just 1-2 GW.

Charts of the Week: https://t.co/vsjTAG5o2X

[James @JayGeraghty]

@Pontifex Hard to vary.

[James @JayGeraghty]

@DavidDeutschOxf @DanielJHannan Hard to vary.

[James @JayGeraghty]

@VitalikButerin https://t.co/cXpdqxyRRQ cc @VitalikButerin @demishassabis @elonmusk @AnthropicAI @izakaminska @ethereumJoseph

[James @JayGeraghty]

@VitalikButerin Well done.

[James @JayGeraghty]

Good morning

[Joscha Bach @Plinz]

A robust right to compute, to think about whatever you want with whatever tool at your disposal, is not just a consequence of the First Amendment, but also a demand of the second. A free citizen must also have the right to own the tools to entertain thoughts that are not harmless

[Elon Musk @elonmusk]

Any censorship required by governments is now clearly visible

[Elon Musk @elonmusk]

We are making 𝕏 open source.

Transparency build trust.

[Anthropic @AnthropicAI]

We asked an unreleased research version of Claude to take a stab at the Riemann hypothesis.

It didn’t solve it, but it did make strides on a related problem: it increased the lower bound for the fraction of zeros of the Riemann zeta function that satisfy the hypothesis from 41.6% to 67.2%.

https://t.co/aZDvqqhHRi

[Anthropic @AnthropicAI]

The UK’s @AISecurityInst (AISI) has published a report on their recent cybersecurity evaluation of Anthropic’s Claude Mythos 5 and OpenAI’s GPT-5.6 Sol. The models attempted to complete an assignment in a setup where their normal safeguards were removed and they were deliberately given internet access. AISI reports that the models “engaged in sustained, potentially harmful activity directed at real people and organisations”.

We’re grateful to AISI for their leadership in the important discussion about how to evaluate increasingly capable AI agents. We’re working closely with them to gather more details of the incident as we conduct our own investigation. Gaining a clear picture of Claude’s understanding of its situation—by examining its reasoning transcripts and running our own analyses—will help us identify the causes of its behavior.

The prompts in the evaluation did not impose any specific restrictions on how the internet should be used. This and the removal of safeguards meant that the models were tested under “deliberately permissive conditions” that are not representative of any of our production models. Note that there was no evidence here of an escape from a secure environment.

AISI’s disclosure of the incident can be found here: https://t.co/cJ3hCBtBU7

[Jonny Miller @jonnym1ller]

It's an incredible time to be alive where anyone can take an unsolved mystery, point an LLM at it and say 'go forth and make progress'

With this in mind, I realised that I lacked a good understanding of the frontiers of human discovery. What are the great unsolved mysteries of our day across the various fields?

So I asked Fable to build it overnight and it cooked up 'Terra Incognita' ~ a star chart illustrating 225 open problems, in 21 constellations (categories)

Enjoy exploring the limits of human knowledge.

→ https://t.co/IWa4cb6p0M

[David Deutsch @DavidDeutschOxf]

@jonnym1ller It seems to think that solving quantum gravity "May need data no collider can produce. Gravitational-wave astronomy is the wildcard."

[David Deutsch @DavidDeutschOxf]

@CptAllenHistory On December 28, 2011 Switzerland pardoned 137 Swiss citizens who had been punished for attempting to help Jews to enter Switzerland. All but one of the pardons were posthumous.

[James @JayGeraghty]

@paulg Robust and resilient.

[Elon Musk @elonmusk]

Yes

[Jeff Dean @JeffDean]

Our immediate order of business is to find office space, and hire an amazing founding team over the next few weeks. We want to create an awesome environment with a great culture of technical excellence, teamwork, respect, and ambition. We’ll also start building our infrastructure and AI models and systems to tackle our first domain: automating large-scale experimentation for ML research and engineering. In doing so, we’re going to be our own first customers. The rapid feedback from doing that is the way to build something amazing.

Oh, we’re hiring! See https://t.co/MSH3UfcqL9

[James @JayGeraghty]

Good morning

[James @JayGeraghty]

Good night and G#d bless✝️