Papers and writing from Zero Drift.https://zerodrift.co.uk/https://zerodrift.co.uk/writing/an-eight-property-review-you-can-challenge/https://zerodrift.co.uk/writing/an-eight-property-review-you-can-challenge/A practical starting rubric for finding work hidden behind a polished interface, and why each property needs a check that can fail.Wed, 16 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/writing/how-this-site-is-checked/https://zerodrift.co.uk/writing/how-this-site-is-checked/The website itself is a useful test of Zero Drift's claim that specifications, mechanical gates and human judgement belong together.Wed, 16 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/writing/the-session-card-is-not-a-prompt-trick/https://zerodrift.co.uk/writing/the-session-card-is-not-a-prompt-trick/A bounded AI build session works because the substance and failure-capable checks are on disk, not because the instruction sounds strict.Wed, 16 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/writing/which-zero-drift/https://zerodrift.co.uk/writing/which-zero-drift/Zero Drift Ltd in the UK and ZeroDrift's US AI enforcement business share a similar name; the entities and services are different.Wed, 16 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/writing/why-staging-is-on-the-product-card/https://zerodrift.co.uk/writing/why-staging-is-on-the-product-card/A working demonstration is useful evidence, but it is not the same as a live service or paying adoption.Wed, 16 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/writing/why-the-rate-card-is-public/https://zerodrift.co.uk/writing/why-the-rate-card-is-public/The price is not a secret weapon. A buyer should know the commercial shape before spending a meeting discovering it.Wed, 16 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/compile-the-specification/https://zerodrift.co.uk/papers/compile-the-specification/AI can reduce the effort of producing code, increasing the relative importance of resolving ambiguity before implementation. Specification-driven development turns business intent into versioned schemas, state, contracts and executable acceptance evidence before agents multiply the wrong assumption.Tue, 15 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/depth-signature/https://zerodrift.co.uk/papers/depth-signature/A feature is not complete because its happy path renders. Its required depth depends on the system it changes. This paper defines a risk-shaped signature for testing security, data, concurrency, recovery, performance, accessibility and operations at the right level.Tue, 15 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/deterministic-gap/https://zerodrift.co.uk/papers/deterministic-gap/Generative AI can propose, explain and build at extraordinary speed. It still cannot make a consequential answer true. This paper separates inference from computation and argues for systems in which models expand the option space while explicit rules, state and evidence close it.Tue, 15 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/five-laws-agentic-delivery/https://zerodrift.co.uk/papers/five-laws-agentic-delivery/Agentic engineering fails when a persuasive session is mistaken for a delivery system. Five operating laws turn model capability into controlled production work by moving plans, units, depth, model choice and parallelism into inspectable artefacts and gates.Tue, 15 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/local-ai-workstation/https://zerodrift.co.uk/papers/local-ai-workstation/Frontier cloud models and controlled local models are complements. Use scarce frontier reasoning to compile specifications and resolve ambiguity; use open-weight models on owned or rented hardware for bounded engineering roles, with harnesses and CI/CD gates deciding what survives.Tue, 15 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/metamorphic-oracles/https://zerodrift.co.uk/papers/metamorphic-oracles/AI may reduce the cost of authoring and maintaining behavioural challenges, giving teams reason to reassess the economics behind the test pyramid. Metamorphic relations, mutation and system-level observation can turn tests from examples of success into an adversarial evidence system.Tue, 15 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/utility-economics/https://zerodrift.co.uk/papers/utility-economics/Software construction and operation are becoming cheaper, but much of SaaS still prices scarcity, lock-in and feature withholding. This paper argues for complete products sold like dependable utilities: low-friction, fairly margined and valuable because they work, not because leaving hurts.Tue, 15 Sep 2026 00:00:00 GMThttps://zerodrift.co.uk/papers/zero-divergence-refactors/https://zerodrift.co.uk/papers/zero-divergence-refactors/Legacy replacement does not require a cut-down successor and a decade of feature-parity work. If behaviour can be observed, specified and challenged, a modern system can be built behind an equivalence boundary and moved into service without forgetting its users.Tue, 15 Sep 2026 00:00:00 GMT