
Provenance beats polish: what shipping document converters taught me about trust
I've shipped a family of document-to-Markdown converters — PDF, DOCX, PPTX, XLSX, EPUB, HTML — and...
Tag archive

I've shipped a family of document-to-Markdown converters — PDF, DOCX, PPTX, XLSX, EPUB, HTML — and...
Why in-toto layout stability and distinct signer identities determine verification value.
A practical evidence model for the NIST manufacturing supply-chain traceability framework.
By Achim Nierbeck and Jamie Goodyear Ask anyone who ships a language model a simple question: which...

I used to treat “who said this” as a display problem. Put my name at the top of an API response, add...
Two 2026 empirical studies land the same shape: tests generated from a task description alone catch roughly 25% of injected faults, while tests generated after the model has seen the implementation catch roughly 14%, and chain-of-thought and chain-of-verification prompting do not close the gap. Red-green-refactor's temporal order was never a workflow preference. It was a provenance claim about the test, and the agentic workflow that generates code and then back-fills tests inherits a measurable disadvantage the human discipline was always built to prevent.

Composite — newsroom war room, election week. A clip lands in Slack with three fire emojis. Someone...
A recycler in Bellevue, Washington, opened one of his donation bins in July expecting the usual: dead...
Why an SBOM without digest-bound signature is a description rather than evidence, and the four verification checks a consumer has to run.
Every x402 settlement you have ever seen was, until recently, opaque. The merchant told you what they...
An agent that can only modify a fixed set of rows has a hard time producing an open-ended...
In August, OpenAI published a post with a title I did not expect to read from a frontier lab this...