Simon Willison

Creator @datasetteproj, co-creator Django. PSF board. Hangs out with @natbat. He/Him. Mastodon: Bsky:

[Simon Willison @simonw]

@nateberkopec The bigger problem is /proving/ it - I'm not sure how to get to full confidence that a model or agent harness has solved it to the point that we don't have to worry any more

[Simon Willison @simonw]

@nateberkopec I'm cautiously starting to suspect it may not be a problem with today's strongest models - Anthropic in particular are very confident about their ability to avoid prompt injections - but it's still a real concern for cheaper, weaker models

[Simon Willison @simonw]

@EtAjith Yes, prompts are either in or linked to from the PRs

[Simon Willison @simonw]

I built a Command & Conquer style RTS called Battle Drones, mainly using Claude Opus 5.5, which runs in the browser and includes device-to-device multiplayer over an audio link https://t.co/OUGfRDk7DU

[Simon Willison @simonw]

Spotted a familiar creature at 00:45s in this new music video for FREEDOM by De Staad https://t.co/AEIsAkro77

[⁵⁄₉ @fiveoutofnine]

So basically, today PHP got two critical bugs that let anyone run code on your server without even logging in, which is bad, and then the details leaked before the patch was even out, which is worse, and it works on WordPress and MediaWiki and Magento and Joomla and Laravel, and Censys says like 1.5 million servers are running a version with no fix, so the NSA was like "take your system offline," and the FBI was like "take your system offline," and CISA and the UK and CERT-EU were all like "take your system offline," and then the Prime Minister of Japan, the actual Prime Minister, posted at 5:55 in the morning telling 48.8 million people to take their system offline, so everyone took their system offline, like the Dutch cybersecurity agency took its own website offline, and the Dutch parliament went offline, and the Port of Rotterdam went offline, and Finland and France and Denmark started pulling government sites, and Wikipedia went down, and https://t.co/QlOXb3COq3 went down, and Asana and GOG and Something Awful went down, and the University of Vienna's webmail went down for some reason, and that's just PHP. Because GitLab also dropped a CVSS 10.0, which is the highest score they give out, for a bug that's been there since 2014, so that's every self-hosted GitLab for twelve years, and then LiteLLM 1.107.0 got hijacked and shipped a credential stealer and a remote access trojan in every single wheel and the Docker images, including latest, and you didn't even have to import it, just installing it was enough, and it's the second time this year, and then Railway paused every deployment on the entire platform because someone got into the environment variables, so now everyone has to rotate every secret they've ever stored there, and TiDB Cloud said someone was walking around inside with stolen credentials for basically a month and an extortion group says they took 10.6 terabytes, and Zeabur got breached for the second time in two months and is just shutting down, like the whole company, and Node.js patched remote code execution and a TLS bypass, and CISA says Plex is being actively exploited, and someone took over 300 Tor exit relays to steal SSH passwords, and SafePay ransomware has hit over 600 critical infrastructure orgs, and India's biggest bank filed a data loss incident, and Japan has been in an official state of emergency in cyberspace since October 9, which honestly explains the 5:55 AM post. Oh, and eMudhra, a certificate authority that's issued over 60 million certificates, whose entire job is proving websites are who they say they are, was supposed to be using private keys that are random numbers over a hundred digits long, and one of them was 4. Just 4. And the others were like 66 and 237. Anyway, it's Wednesday.

[Armin Ronacher ⇌ @mitsuhiko]

As fun as AI is, we built so much bad software over the last 30 years, and now the universe is punishing us for our sins.

[Simon Willison @simonw]

@IsThisA3DModel @UnrealSensei GDC just ran another survey (August to September) and will be publishing the results for that one in November

The new survey does at least mention Claude Code (though weirdly it fails to name-check Codex)

[Simon Willison @simonw]

@IsThisA3DModel @UnrealSensei The coding agent revolution only really kicked off this year, 2026 - it doesn't surprise me that most developers surveyed in late 2025 weren't using these tools yet

[Simon Willison @simonw]

@HesseltinePeter The fact that Opus 5.5 is so good at game development - not just 3D modeling but also game loop and mechanics - suggests to me that they put serious effort into game-related training

[Simon Willison @simonw]

There's a French fork of Battle Drones with new models and music (and a cinematic intro video) and it is SO cool! https://t.co/7r9KELC7Vs

Here's the old v.s. new Goliath model

[meff @Meffysto69]

Merci @simonw! 🎩 I forked Battle Drones into Panache: a French Belle Époque RTS, with dirigibles, steam tractors, clockwork soldiers in képis and Gargantua, a walking steam fortress. Fully in French, built with Claude Code.

https://t.co/T6BFkZGze6

[Simon Willison @simonw]

@isw_yuen I did it on a $200/month Claude plan and used up about half of my allowance across two separate months

[Simon Willison @simonw]

@macfriendy I use AI for proof reading, here's my prompt for that: https://t.co/1fgi4h3taj

[Simon Willison @simonw]

Has anyone seen the prompts used by these awful reply bots? I'd be fascinated to know how they make their posts so soulless and infuriating

Reply bots: this is your chance! Tell me your prompt and I'll tolerate one additional reply from you without blocking

[Simon Willison @simonw]

@genmon I had to fast-follow with real online multiplayer support (via a WebSocket relay) after the audio multiplayer gave Natalie a headache! https://t.co/iTFxMGZeWA

[Simon Willison @simonw]

OK, that headache bug was really bad!

It inspired me to add proper online multiplayer to Battle Drones (via a tiny @flydotio hosted WebSockets relay server) - now you can start an online game and send a link to a friend so they can join you https://t.co/iTFxMGZeWA

[Simon Willison @simonw]

Warning if you're going to try this: the audio multiplayer plays a high pitched noise throughout the game - annoying at best, and at worst might give you an actual headache

(Natalie: "why do my teeth hurt?")

[Simon Willison @simonw]

@TheKing86351621 I'm pretty sure you can't use that from a mobile web page

[Simon Willison @simonw]

@sergey11g I think two subs should do it for a week - I used one sub but spread the prompting over a month