The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 44 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
Willison: GPT-5.1 and Claude Opus 4.5 marked an inflection point for coding agents
“In November, we had what I call the inflection point where GPT, 5.1 and Claude Opus 4.5 came along and they were both just, they were incrementally better than the previous models, but in a way that crossed a threshold where previously, if you had these coding…”
Willison: Code is easier for AI agents because it is verifiably right or wrong
“Like code is easier than almost every other problem that you pose these agents because code is obviously right or wrong. Like it produces code. You run the code. Either it works or it doesn't work. There might be a few subtle hidden hidden bugs, but generally …”
Willison: Large-scale agentic engineering will always require deep developer expertise
“The art of having them help you build software, you could deploy to a million people. That's not, that's never going to be easy. That's never going to be trivial. That's always going to require a great deal of depth of experience in what software and how softw…”
Willison: StrongDM implemented a policy where engineers never read code
“The next rule though is nobody reads the code and this is the thing which StrongDM started doing back in, I think it was August last year. They said, okay, we're not going to read the code.”
Willison: AI agents have become credible security researchers in recent months
“At the same time, the agents are getting really good at security penetration testing now. And this is a new thing. I think in the past, again, in the past sort of three to six months, they've started being credible as security researchers, which is sending sho…”
Willison: Mid-career software engineers face the greatest AI displacement risk
“The problem is the people in the middle. Like if you're mid career, if you haven't made it to sort of super senior engineer yet, but you're not sort of new either. That's the group, which ThoughtWorks resolved work probably in the most trouble right now.”
Willison: AI can never have true agency without human motivations
“I think agents have no agency at all. Like I would argue that the one thing AI can never have is agency because it doesn't have human motivations. Sure. You can tell it make more money or whatever, but it's never going to be able to decide on its, like what ma…”
Willison: AI broke tests and documentation as quality signals
“Like it used to be, if you looked at software and it had high quality tests and documentation and everything, it meant it was good. And now that signal is gone.”
Willison: AI-written code will become common for engineers by end of 2026
“I expect by the end of this year, it will not be uncommon to have an engineer say that almost all of that code is written by AI.”
Willison: Coding agents become truly productive only in permissionless 'unsafe' mode
“I think a lot of people who haven't got on board with coding agents yet, haven't tried them in the unsafe mode. They're using coding agent where it's like, oh, can I run this piece of code? Can I edit this file? And that means you have to pay complete attentio…”
Willison: Pelican SVG quality strongly correlates with general LLM capability
“There appears to be a very strong correlation between how good their drawing of a pelican riding a bicycle is and how good they are at everything else. And nobody can explain to me why that is. But as I started looking at these things, I realized, wow, The bet…”
Willison: AI coding agents make verbose, extensive test suites economical
“If you look at a repo and there's huge amounts of tests that aren't really doing anything interesting, that's really expensive because now when you change the code, you've got to update a thousand lines of tests and all of that.
It turns out I don't care anymo…”
Willison: LLMs fundamentally cannot separate trusted instructions from untrusted user text
“Agents fundamentally, like LLMs, can't tell the difference between texts that you give them and texts that you copy and paste in from other people. They're all the same thing. So instructions in that input text can always override the earlier instructions.”
Willison: AI will eventually suffer a catastrophic Challenger-style security disaster
“So my prediction is that we're going to see a challenging disaster. Like at some point, this is going to catch up with us and it's going to be Very, very, very bad, and that will hopefully help us start trying to figure out how not to do this. At the same time…”
Willison: OpenAI and Anthropic didn't build OpenClaw due to security risks
“The reason OpenClaw took off is Anthropic and OpenAI could have built this and they didn't because they didn't know how to build it securely. If you're an independent third party, you don't have that restriction. You can just Build something and put it out the…”
Willison: Vibe coding without reviewing code is irresponsible for multi-user software
“If you're vibe coding something for yourself, where the only person who gets hurt, if it has bugs, is you go wild. That's completely fine. The moment you did your vibe coding code for other people to use where your bugs might actually harm somebody else. That'…”
Willison: StrongDM spent $10K daily on tokens simulating software testers
“Like they were spending 10,000 dollars a day on tokens, I think, simulating all of these end users. I believe so. But it meant that their software was being very robustly tested in all of these different ways.”
Willison: StrongDM used coding agents to simulate Slack, Jira, and Okta APIs
“So what they did is they built their own simulation of Slack and Jira and Okta and all of this software they were integrating with. And the way they did that is they basically took the API documentation for the public APIs for Slack and the client libraries, t…”
Willison: OpenAI and Anthropic maintain private, restricted security models
“Both OpenAI and Anthropic have specialist security models That they will not release to the general public because they can be used to break into websites. So they have like invite only, like registered security researchers can apply for access and they've bee…”
Willison: Simulated AI user testing cannot match real human usability feedback
“I don't think that's credible. I don't think you're going to get as good results from ChatGPT pretending to click around on your prototype than you would from an actual human being.”
Willison: Agentic coding removes developers' need for multi-hour focus blocks
“People talk about how important it is not to interrupt your coders, right? Your coders needs to have Like solid two to four hour blocks of uninterrupted work so they can spin up their mental model and tune out the code. It's some, that that's changed completel…”
Willison: Coding agents search local repositories to synthesize relevant code
“Coding agents can do searches. So you can give them access to an entire hard drive full of stuff and tell them what you need to solve, and they will run search tools to find just the examples that they need to piece things together.”
Willison: Abandoning automated tests with AI coding agents is a huge mistake
“I think it's a huge mistake if you drop tests in exchange for speed of development, because
Very quickly when you're working the test, you find your development speed goes up.
The existence of the test lets you move faster because you don't have to constantly …”
Willison: Minimal code skeletons guide AI agents better than long Claude.md instructions
“So sometimes, some people will tell you should have a Claude.md with like paragraphs of text describing how you like to work. I don't tend to do that because instead I start with a very thin skeleton that just gives it enough hints on how I like to work that i…”
Willison: AI 'lethal trifecta' risk is solvable only by eliminating one core capability
“You have a lethal trifecta. Anytime your agent has three things, it's got Access to private information. There's information that you've exposed to it, like your private inbox that, that is, is private in some way. It's exposed to malicious instructions. So th…”
Willison: AI prompt injection benchmarks under 100% provide false security
“And again, until it's a hundred percent, I don't think it's a meaning. I think it just gives people a false sense of security that this problem won't bite them.”
Willison: Building a secure OpenClaw is AI's biggest current opportunity
“So I think the biggest opportunity in AI right now, if you can build safe OpenClaw, if you can deploy a version of OpenClaw that does all the things people love about it and won't randomly leak people's data and delete their files, that's a huge opportunity.”
Willison: Anthropic and OpenAI focused 2025 training efforts on coding
“Both Anthropic and OpenAI spent the whole of 2025 focusing all of their training efforts on coding.”
Willison: All software development will eventually be mediated through AI
“We're all moving in the direction where
Our code is mediated through AI at some point.”
Willison: AI generates roughly 95% of the code I produce
“And today, probably 95% of the code that I produce, I didn't type it myself.”
Willison: Anthropic found and responsibly disclosed 100 vulnerabilities to Mozilla
“I think Firefox just a few days ago, maybe last week said that they'd, they'd done a release, which was assisted by Anthropic. Anthropic had Discovered a hundred like potential vulnerabilities in Firefox and responsibly reported them to Mozilla who then fixed …”
Willison: Product designers not vibe coding prototypes miss AI's biggest boost
“I think anyone who's doing product design isn't vibe coding little prototypes is missing out on the latest, but like the most powerful sort of boost that we get in that step.”
Willison: Handcrafted software will be valued more in the future
“Artisanal to handcrafted software, I think is going to be valued more.”
Willison: AI makes rapid prototyping almost free, democratizing the skill
“Prototyping is almost free, I think. And that really impacts me because throughout my entire career, my superpower has been prototyping. Like I am very, I've been very quick at knocking out working prototypes of things. I'm the person who can show up at a meet…”
Willison: LLM search integration is now better at searching than humans
“Now that all of the major models have really good search integration, they're just better at searching than I am. I can ask them a question and watch them fire off five searches in parallel for like aspects of answering that question, pull the data back.”
Willison: Practices helping AI agents code also improve human engineering
“Most of the things that make agents write better code work for humans too.”
Willison: Prompting AI agents with 'red/green TDD' yields better code
“If you get them to write the tests first.
You do get better results because they're much less likely to forget to test something or to add bits of code that aren't necessary.
And so you could tell them,
Write this using test, make sure that you write the test …”
Willison: Coding agents excel at adopting existing patterns from single example files
“And the reason for this is it turns out coding agents are phenomenally good at sticking to existing patterns in the code. Like if you give them a code base that already has just a single test in it, they will write more tests. They will notice that if you've g…”
Willison: Human-in-the-loop AI safety requires filtering requests to high-risk actions only
“Human in the loop helps a little bit, but if you ask the human to click, okay, five times a minute, they'll just click. Okay. All the time. If you can filter it down. So the human only gets asked on the high risk activities. That's how you build a sort of A pe…”
Willison: Building a 'claw' will be AI engineering's Hello World
“So like, I think the new hello world of AI engineering is going to be building your own claw.”
Willison: Journalists are better equipped for AI than most professions
“The art of journalism is you talk to a bunch of people and some of them lie to you and you figure out what's true. So as long as the journalist treats the AI as yet another unreliable source, they're actually better equipped to work with AI than most other pro…”
Willison: Cloudflare and Shopify planned 1,000 interns due to AI onboarding
“Cloudflare and Shopify, both said they were hiring a thousand interns over the course of 20, 25. Because the intern onboarding costs, it used to be takes a month before you enter and can do anything useful. Now they're doing something useful within like a week…”
Willison: OpenClaw progressed from first commit to Super Bowl ad in months
“The first line of code for OpenClaw was written on November the 25th. And then in the super bowl, there was an ad for AI.com, which was effectively a vaporware white labeled open claw hosting provider. So we went from first line of code in November to super bo…”