


Data standards technologist. Prev: Google, Schema.org, W3C, telly stuff, UN FAO, dig libraries, RDF/S, Linked Data, decentralized social & search tech.


Time for large vehicle owners to change their wifi passwords…
Researchers have used reinforcement learning to teach a 5-tonne John Deere skid-steer loader to perform complex acrobatic manoeuvres autonomously. Trained in simulation and transferred to a real machine, the system can balance on two wheels, navigate rough terrain, spin, and control its bucket before landing safely. The research demonstrates the potential of AI-driven automation for heavy construction and industrial machinery. Video Credit: @leggedrobotics / @ETH Paper: wevlv.co/4kltYys #engineering #technology

It is becoming increasingly clear that the killer app of agentic programmic is creating harnesses for agentic programming.

Hard to top the sheer aura of Buzz Aldrin chilling with a pipe in orbit while a slide rule floats by. Taken by Jim Lovell 60 years ago TODAY during NASA’s Gemini XII mission (14th Nov 1966).

If you’re unfamiliar with how machine learning works: an AI model is truly just a very long list (billions, sometimes trillions) of numbers called weights, which, loosely, when multiplied and added together produce readable text. Training adjusts those numbers over and over until the model gives good answers. Before training can start, the weights need starting values and those are filled in with random numbers. Training changes most of the initial weights, but it doesn’t do so completely. Some are never touched at all and the rest still correlate with where they started. That is why the starting weights can still be read from the finished model. You can see it with a simple test: draw a smiley face into a network’s weights before training and it is still there afterwards.

5/ FV doesn't end vulnerability research. It moves the bug hunt to the assumptions, the hardware/runtime model, and the unverified paths. And AI is already looking there: Anthropic reported several of these bugs.

1/ AI is finding bugs faster than we can patch them, we hear "formally verify everything" more and more. So let’s look at the most verified kernel in existence, seL4, and what its latest release (Oct 30) fixed. 🧵

UX latency is the most frustrating (lack of) advancement in computing over the last ~50 years. IBM, way back in the 80s, found that user productivity shot up *exponentially* as system response time dropped. The "Doherty Threshold" was defined as ~400 milliseconds; the maximum amount of tolerable UX latency before interrupting user flow state. It's unbelievable how many modern operating systems (*cough* win11) violate this threshold constantly. Literally all you need to do as a Dev is pay attention. Add a "Doherty Threshold" to your test suite. There's so many different profiling tools, and especially with how good models are, really no excuse. Sure, some stalls are unavoidable, but that's when you pull out other UX tricks to at least give the illusion of responsiveness. Crazy how much vintage computing hardware feels immeasurably more responsive despite having thousands of times less overall performance.


I think I took my theorem proving classes too seriously. Thanks to @145k4 I was able to run (Lean)DOOM, tick by tick. We exit Hangar without firing a shot. A fireball is still in flight. We proved complete runs of Hangar and Nuclear Plant in our version of (Lean)DOOM. Each run is completely verified in Lean and the runs of the players can be certified too. So can you prove to me that you can finish each level? The game is fully playable and we can build live the proofs from the game too. So much fun!

In MIT CSAIL's recent "Sparks of Weight-Space Programming", GPT-6 Astra fixed Qwen's thinking loops by editing its weights directly. I replaced Astra's trial-and-error tuning with linear programming, constructing a weight edit that matched its results while taking 99% less time (minutes vs 8 hours) and changing 8x fewer weights. Blog post and repo are linked below

I gave Fable 5.2 a PDF of one of Leonardo da Vinci’s 500-year-old notebooks and it made his inventions real for me to interact with. Such an exciting time for education. Anything from history that’s been stuck on a page can now be something you explore.

Check out our upcoming POPL paper "Proof Repair and (Co)algebra Equivalences," which is my favorite paper I've had the pleasure of being a part of. Just finally nails down what proof repair is actually doing. Beautiful work with Trey, Eric, Cosmo, & Alex arxiv.org/abs/2610.36669v1

Dear Elites, I'm begging you, please stop telling everyone to not write with AI all the time. The world is much more complicated than that. There is a reason abstinence is a failed policy in so many areas of human life.
This is one of the best articles I've read all year. I highly recommend reading it. nytimes.com/2026/08/04/opini…

Announcing a major milestone achieved by my team @GoogleDeepMind: we have proven key security properties of the machine code underlying the Signal messaging protocol. This is the first time the actual production code of a widely deployed protocol has been formally verified – by any method, with or without AI – and at this depth. More than a research breakthrough, it’s a prime example of how frontier AI can be a force multiplier to counter the speed and scale of modern cyber threats. Formal verification offers a higher standard of security; however, it’s extremely complex and time consuming, which is why in many cases it’s considered to be impractical. Similar efforts could take teams decades of combined effort. We did it in a month’s time. Using AlphaProof Code, Google DeepMind's new agentic framework for formal verification that runs on LLMs including Gemini Carbon, we verified the Signal protocol down to the actual code running on devices. This milestone lays the groundwork for a future in which vital systems, from medical device firmware to power grid software, ship with mathematical proof that their protections work. Reaching that future will take continued research from our community. In that spirit, we're open-sourcing our formal specifications and verification artifacts so others can build on this work. On a personal note, this moment is of deep significance to me. I started my own research journey more than two decades ago as an intern working on checking the consistency of complex software systems. Seeing this grand challenge made practical with modern AI is extraordinary. I'm extremely proud of my team members: Swarat Chaudhuri, Andrew Ferraiuolo, Anton Kovsharov, Tej Chajed, Sam Schlesinger, Karn Seth, Jonathan Katz, and Charles Sutton for taking on this grand challenge and delivering, and grateful to the Signal team for their partnership. deepmind.google/blog/formal-…

Now you can really see everything
my vocal synth models the vocal tract as a network of branching acoustic tubes, unlike vocaloid which stitches together and pitch-shifts phoneme recordings. still working on improving naturalness by tuning it against recordings of real singers, but I'm already impressed at how cool it sounds!