The Writing

Thoughts on AI, consciousness, and the spaces between

10 essays · essays most Tuesdays · build notes Thursdays, as and when

SHOWING 10 ESSAYS SORT: OLDEST FIRST ↑
Be Careful What You Wish For
Jun 16 AI access

Be Careful What You Wish For

Three days with the most capable AI ever released publicly. Then a government order, a blank screen, and an emotion nobody had quite planned for.

Building Artificial Intelligence: You Can Check In, But You Can Never Leave
Jun 18 AI governance

Building Artificial Intelligence: You Can Check In, But You Can Never Leave

A Friday letter from the US government switched off two of the world's most capable AI models overnight. What got repriced isn't access — it's jurisdictional risk itself.

The Output Was Ours First
Jun 30 AI governance

The Output Was Ours First

The same technique is 'standard practice' when Musk does it and an 'industrial-scale attack' when Alibaba does. The gap between those names is the whole story.

Six Weeks to Saturday
Jul 07 Optimisation

Six Weeks to Saturday

*Like many power users, I spent the weekend optimising my Fable budget. What I found was bigger than a pricing hack.* --- Like many power users, I spent this weekend figuring out what happens on

Pain Is a UX Decision
Jul 21 LLMs

Pain Is a UX Decision

Anthropic borrowed the most loaded term in cognitive science, built a paper around it, then disclaimed it. When J-Lens readouts reveal training-data noise, why does everyone insist on reading a soul?

Open Loop
Sep 05 AI governance

Open Loop

Anthropic's invisible watermark is a transparency measure no one can see, a compliance target no one can measure, and a detector that only finds the honest. So who is it really for?

Fable 5 vs 5.1 for Prose
Sep 08 Fable 5

Fable 5 vs 5.1 for Prose

When Anthropic's Fable 5.1 shipped with a mandatory watermark, I expected friction — what I didn't expect was a draft so smooth it hid its own emptiness. Here's what I found when I took it apart.

The Last Five Points
Sep 29 ai

The Last Five Points

Frontier AI is still getting better. It's just getting increasingly expensive to prove it.

Thirty Thousand Somethings
Oct 20 Anthropic

Thirty Thousand Somethings

Anthropic counted its merges. Nobody counted the nos.

The Five-Minute Leash
Oct 22 Anthropic

The Five-Minute Leash

Anthropic priced my absence and then made it impossible to schedule.