↵ select ↓ ↑ navigate esc close

Simon Willison's Weblog

npub1cgctsayk4d52dfauphf94muyxm8fl9526mf50gdllgaf4agh6clqgexk9p@drss.io

Token fraud is here and it’s BAD (YouTube)

20 Nov 2026

[**Token fraud is here and it’s BAD (YouTube)**](https://www.youtube.com/watch?v=ZDpP7JXPlhc) Claire Vo runs [ChatPRD](https://www.chatprd.ai), an AI platform that helps product managers "turn an

PostgreSQL 19 Released

19 Nov 2026

[**PostgreSQL 19 Released**](https://www.postgresql.org/about/news/postgresql-19-released-3449/) Most years PostgreSQL ships a new release in September, but this year's release came two months later

llm-gemini 0.38

18 Nov 2026

**Release:** [llm-gemini 0.38](https://github.com/simonw/llm-gemini/releases/tag/0.38) Today Google [introduced Gemini 4

datasette-arrows 0.1a0

18 Nov 2026

**Release:** [datasette-arrows 0.1a0](https://github.com/datasette/datasette-arrows/releases/tag/0.1a0) [Datasette 1.0a43](https://docs.datasette.io/en/latest/changelog.html#a43-2026-11-15) added

Quoting Daniel Stenberg

18 Nov 2026

> During the time these agents converted 185,000 lines of C code into 340,000 lines of Rust code (I counted), we merged over 300 commits in the actual curl project. We actually improved curl during

The PSF end-of-year fundraiser

18 Nov 2026

[**The PSF end-of-year fundraiser**](https://donate.python.org) With my Python Software Foundation board director hat on: we've launched our end-of-year fundraiser for the non-profit that supports

External sandboxes for coding agents

17 Nov 2026

[**External sandboxes for coding agents**](https://blog.primeradiant.com/external-sandboxes-for-coding-agents/) I've been exploring new ways to sandbox code that is executed by agents, with a focus

GPT‑6.2 Astra and GPT‑6.2 Sol

17 Nov 2026

[**GPT‑6.2 Astra and GPT‑6.2 Sol**](https://openai.com/index/introducing-gpt-6-2-astra-sol/) OpenAI: > We’re introducing GPT‑6.2 Astra, our new frontier model for your hardest projects, and

Running the full server-side Datasette Python app in the browser with service workers and Pyodide

17 Nov 2026

[Datasette Lite](https://lite.datasette.io/) is my port of the [Datasette](https://datasette.io/) Python application to run directly in the browser using WebAssembly and

Quoting Go Proverbs

17 Nov 2026

> A little copying is better than a little dependency. — [Go Proverbs](https://go-proverbs.github.io), [via](https://news.ycombinator.com/item?id=50516689#50526408) Tags:

llm 0.37.3

16 Nov 2026

**Release:** [llm 0.37.3](https://github.com/simonw/llm/releases/tag/0.37.3) A bug fix release for [LLM](https://llm.datasette.io), mainly to cover a problem with the most recent Homebrew package of

datasette-maplibre 0.1a0

15 Nov 2026

**Release:** [datasette-maplibre 0.1a0](https://github.com/datasette/datasette-maplibre/releases/tag/0.1a0) The first new Datasette geospatial plugin in quite a while! This one combines the

datasette 1.0a43

15 Nov 2026

**Release:** [datasette 1.0a43](https://github.com/simonw/datasette/releases/tag/1.0a43) I've been leaning hard into Claude Fable 5.2 to help knock things off the [1.0

datasette 0.65.6

15 Nov 2026

**Release:** [datasette 0.65.6](https://github.com/simonw/datasette/releases/tag/0.65.6) I backported a minor security fix from 1.0a43: > Fixed a security issue where a database registered using

datasette-llm 0.1a11

14 Nov 2026

**Release:** [datasette-llm 0.1a11](https://github.com/datasette/datasette-llm/releases/tag/0.1a11) `datasette-llm` is the intermediate plugin for Datasette that supports configuring LLMs for use

datasette-vega 0.7

14 Nov 2026

**Release:** [datasette-vega 0.7](https://github.com/simonw/datasette-vega/releases/tag/0.7) A new release of Datasette's longest untouched plugin, for rendering charts (using

datasette-cluster-map 0.19

13 Nov 2026

**Release:** [datasette-cluster-map 0.19](https://github.com/simonw/datasette-cluster-map/releases/tag/0.19) One of the oldest Datasette plugins, and still one of my favorites. It now works with

datasette-leaflet 0.3

13 Nov 2026

**Release:** [datasette-leaflet 0.3](https://github.com/simonw/datasette-leaflet/releases/tag/0.3) OpenStreetMap have a very reasonable policy that hits to their hosted tile server [should send a

llm-gemini 0.37

12 Nov 2026

**Release:** [llm-gemini 0.37](https://github.com/simonw/llm-gemini/releases/tag/0.37) Google launched [Gemini 4

Gemini Decisions API

11 Nov 2026

[**Gemini Decisions API**](https://blog.google/innovation-and-ai/technology/developers-tools/gemini-decisions-api/) Google join [Jev](https://simonwillison.net/2026/Sep/21/jev/) and

aws-emulator

11 Nov 2026

[**aws-emulator**](https://github.com/aws/aws-emulator) Finally, an official emulator from AWS to run tests against a local clone of their many different services without having to provision a paid

Browse tools.simonwillison.net by tag

11 Nov 2026

[**Browse tools.simonwillison.net by tag**](https://tools.simonwillison.net/#tags) I've been wanting a better way to browse my [collection of vibe-coded tools]() (244 and counting) for a while. I

llm 0.37.2

11 Nov 2026

**Release:** [llm 0.37.2](https://github.com/simonw/llm/releases/tag/0.37.2) Fixes a security issue reported by [Daniel de Araujo](https://github.com/BluePlexus) in

Anthropic's IPO roadshow video

11 Nov 2026

[**Anthropic's IPO roadshow video**](https://ipo.anthropic.com/us/#roadshow-video) Anthropic kicked off their IPO roadshow yesterday, and their roadshow video has some previously unreleased

llm-gemini 0.36

10 Nov 2026

**Release:** [llm-gemini 0.36](https://github.com/simonw/llm-gemini/releases/tag/0.36) Gemini 4 Carbon was announced [two weeks

Introducing Our Smallest, Most Efficient Model: Muse Ember

10 Nov 2026

[**Introducing Our Smallest, Most Efficient Model: Muse Ember**](https://research.meta.ai/blog/muse-ember) Meta AI just launched their entry in the Claude Haiku/GPT Luna category: Muse Ember, which

Quoting Prompting Claude Fable 5.2

10 Nov 2026

> Claude Fable 5.2 responds closely to what the prompt says about who owns the code it works on. In Anthropic’s testing on small coding tasks, when a prompt said nothing about ownership or

llm-anthropic 0.31

9 Nov 2026

**Release:** [llm-anthropic 0.31](https://github.com/simonw/llm-anthropic/releases/tag/0.31) Anthropic [released Fable 5.2 and Mythos 5.2](https://www.anthropic.com/claude-fable-and-mythos-5-2)

Locating the source of a mysterious beep with Claude

9 Nov 2026

Our house has a box room, full of boxes that we still haven't unpacked from our last move. A couple of weeks ago something in that room started *intermittently beeping*. The beeps were far enough

The Web Is More Than Enough, by Paul Ford

9 Nov 2026

[My comment](https://news.ycombinator.com/item?id=50416456#50421091) on [The Web Is More Than Enough, by Paul Ford](https://news.ycombinator.com/item?id=50416456) — Hacker News. OK, this comment

Hide your slop PR descriptions behind <details><summary>

9 Nov 2026

Most of the changes I make to my projects these days come in the form of PRs opened for me by Claude Code or Codex. The PR is my unit of work - I use them to bundle commits together, iterating on

parse_number() as a SQLite extension: C vs Rust

8 Nov 2026

**Research:** [parse\_number() as a SQLite extension: C vs Rust](https://github.com/simonw/research/tree/main/sqlite-parse-number-c-vs-rust#readme) I was out walking the dog when I got curious about

datasette #3046: Don't serve link: rel=next HTTP headers to Firefox for HTML pages

7 Nov 2026

[**datasette #3046: Don't serve link: rel=next HTTP headers to Firefox for HTML pages**](https://github.com/simonw/datasette/issues/3046) I was looking at the Firefox network panel today when I

Quoting Nilay Patel

6 Nov 2026

> The products aren't good enough to sell people on the costs. > > You cannot tell people that you have to listen to the whine of a data center in their backyard for AI overviews and some of the

Quoting Paul Ford

5 Nov 2026

> The web browser is very, very, very powerful software for applications, *and the cost of accessing that power is going down seemingly every day*. — [Paul

datasette-llm 0.1a10

4 Nov 2026

**Release:** [datasette-llm 0.1a10](https://github.com/datasette/datasette-llm/releases/tag/0.1a10) Fixes a [bug](https://github.com/datasette/datasette-llm/pull/11) that affected my

llm 0.37.1

3 Nov 2026

**Release:** [llm 0.37.1](https://github.com/simonw/llm/releases/tag/0.37.1) A bugfix release, with contributions from [Steven Huang](https://github.com/remote-controlled-man), [Tao

Qwen 3.8 27B on my laptop can interpret model weights, too

3 Nov 2026

Yesterday's hot paper was [Sparks of Weight-Space Programming: LLMs can natively read &amp; write neural network weights](https://rikiyat.github.io/blog/weight-space-programming/) by Rikiya Takehi,

I discovered a sound that leaks your exact location (and it's everywhere) - YouTube

3 Nov 2026

[**I discovered a sound that leaks your exact location (and it's everywhere) - YouTube**](https://www.youtube.com/watch?v=UytYrhYLOKI) Genius piece of OSINT by Colsto, who demonstrates how you can

October sponsors-only newsletter

3 Nov 2026

I sent out the October edition of my [sponsors-only monthly newsletter](https://github.com/sponsors/simonw/). If you are a sponsor (or start a sponsorship now) you can [access it

llm 0.37

2 Nov 2026

**Release:** [llm 0.37](https://github.com/simonw/llm/releases/tag/0.37) There's a *lot* of stuff in this new release of my LLM CLI tool and Python library, representing around two months of

The U.S. Census Bureau and “Super Intelligence”

2 Nov 2026

I was reminded today of [AI Adoption Rate Trending Down for Large Companies](https://www.apolloacademy.com/ai-adoption-rate-trending-down-for-large-companies/), a chart produced by Apollo Global

If in doubt, delete the uv.lock file

1 Nov 2026

I've been getting the occasional confusing Python dependency error since switching over to Python 3.15 and the rule I'm having to remember is: **If in doubt, delete the uv.lock file** Most of my

Quoting Kyle Kingsbury

1 Nov 2026

> More broadly, the use of LLMs erodes trust in documentation, in software, and in the organizations which produce them. Perhaps engineers will begin to treat any new software project with

How We Took Over ChatGPT Accounts

31 Oct 2026

[**How We Took Over ChatGPT Accounts**](https://www.parameter.ai/labs/we-found-a-gap-at-openai) Parameter Labs describe an exploit they found against ChatGPT's implementation of the artifacts

dbreunig/nesformer

30 Oct 2026

[**dbreunig/nesformer**](https://github.com/dbreunig/nesformer) Inspired by [qwenputer](https://divagr.com/blog/qwenputer) by Divyansh Agrawal, Drew Breunig baked a working emulator for a Nintendo

API credits benefit for eligible ChatGPT subscriptions

29 Oct 2026

[**API credits benefit for eligible ChatGPT subscriptions**](https://help.openai.com/en/articles/20001626-api-credits-benefit-for-eligible-chatgpt-subscriptions) Earlier this month Anthropic

Building a multiplayer game world with autonomous Fable-class models

29 Oct 2026

[**How I Built the Most AI-Pilled Company in the World**](https://twitter.com/jsnnsa/status/2115623382721126826) Don't be put off by the clickbait headline, this piece by Jacob Sansbury describing

Quoting Ben Hylak

29 Oct 2026

> every product i use should just be a headless mcp, that feeds into an agent, where i never have to think about anything underneath. > > except for our product, of course. > > in our product,

Quoting Ted Sanders

28 Oct 2026

> In general, LLM performance is incredibly multidimensional. There's no guarantee of a fixed ordering where the model that is best at math is also the model that's best at jokes. Despite that