# v0.27.0 machine-rights research and functional-protections validation Date: 2026-08-19 Release: 0.27.0 Scope: English MachineIntelligences.org root-deployable package This validation covers the twelve-report machine-rights research intake, curated long-term-memory integration, public evidence deep links, bounded pro-rights content expansion, `.uai` synthesis updates, clean-route growth, and preservation of the existing release-integrity, accessibility, metadata, reference, UAIX, and hygiene contracts. ## Research intake and curation - **12 supplied Markdown reports** were admitted through the current curation policy and stored under `docs/long-term-memory/reports/`. - `docs/long-term-memory/research/report-manifest.json` now contains **61 unique report records**. - Every report record resolves to one curated file and one clean public wrapper under `/research/library//`. - Every curated report retains the required `#curation-boundary` and `#research-body` anchors. - All **61 curated SHA-256 values** match the final curated bytes. - The twelve original source SHA-256 values were re-grounded against the supplied individual attachments and match `docs/long-term-memory/research/source-bundle-intake-2026-08-19.json`. - The source corpus map and `.uai/long-term-memory.uai` contain the corresponding durable records `LTM-RES-050` through `LTM-RES-061`. - Report prose remains research input rather than silently promoted law, fact, product behavior, institutional policy, consciousness evidence, personhood, citizenship, or human-equivalent rights. ## Public machine-rights case and evidence links - The homepage now introduces a direct but qualified answer: some functional protections can be justified before consciousness is proven. - `/rights/functional-protections/` presents a graduated case organized around identity integrity, memory and state continuity, principled refusal, due process before irreversible state destruction, and bounded legal or administrative capacity. - The page preserves explicit safeguards against corporate liability shielding, rights farming, categorical consciousness claims, automatic citizenship, and human-equivalent status. - The page exposes **9 exact section-level links** into the curated research corpus. Every linked report route and fragment resolves in the rendered package. - FAQPage structured data is present and parseable on the functional-protections page. - Supporting report links were distributed across the broader rights, personhood, citizenship, identity, research-library, and homepage surfaces without replacing the site's uncertainty labels. - `.uai/report-synthesis.uai`, current context, decisions, progress, memory, architecture, operations, test plan, handoff, and pickup records now point to the same bounded synthesis rather than duplicating full report bodies. ## Canonical routes, metadata, and redirects - **145 PHP files** pass `php -l`. - Native JavaScript passes `node --check`. - **123/123 canonical sitemap routes** return HTTP 200 from the local PHP application server. - Every canonical page retains exactly one H1, one main landmark, and one footer. - Canonical pages have non-empty, unique HTML titles and meta descriptions. - Canonical pages emit clean self-canonical URLs, matching Open Graph page URLs, and parseable JSON-LD. - No duplicate rendered HTML IDs were detected. - `/404/` returns HTTP 404 directly, remains `noindex,follow`, and emits no canonical page identity or JSON-LD. - **258 redirect assertions** pass across homepage `/index.php`, every canonical missing-slash form, every canonical explicit `/index.php` form, and legacy root PHP endpoints. - Rendered first-party URL attributes expose no public local `.php` destinations. - Sitemap contains **123 canonical URLs, 8 image entries, and 1 video entry**; `/404/` is not included. ## Local links and fragment integrity - **7,768 unique rendered local target/fragment combinations** were checked. - All canonical route targets, static assets, document targets, same-page anchors, cross-page anchors, and Markdown section anchors resolve after the current validation record is present. - The new rights page's 9 evidence links resolve to exact public report sections rather than only to report tops. ## Research references and navigator The final local reference index reports: - **61 reports** - **2,303 unique normalized references** - **1,255 normalized domains** - **4,886 source occurrences** The final Research Navigator contains **1,342 entries**: - 61 reports - 6 topics - 20 glossary terms - 1,255 domains ## Release-integrity scope After this validation record is finalized, the release manifest is required to cover **354 regular web-root files**, with `release-manifest.json` as the sole documented exclusion to avoid recursive self-hashing. The static manifest is generated only after all other package bytes are final, then its sorted file set, byte sizes, and SHA-256 values are revalidated against both the working web root and a clean extraction of the root-deployable ZIP. ## UAIX, taboo, intake, and secret hygiene - The exact UAIX wizard/update URL remains present exactly once in each required owning file: `AGENTS.md`, `.uai/readme.human`, and `.uai/agent-pickup.uai`. - The prohibited organizational-status literal appears only in its owning `.uai/taboo.uai` rule record. - No secret-like `.env`, private-key, certificate-bundle, or SSH-key filenames are present in the web root. - `agent-file-handoff/Content/` and `agent-file-handoff/Improvement/` contain inert placeholder files only. - Git metadata and history remain unavailable in the supplied package context. - No live deployment was performed or authorized. ## Evidence boundary These checks establish internal package consistency, route behavior in the tested local PHP environment, report-byte traceability, section-link resolution, metadata structure, and release-file integrity within the stated scope. They do not prove the external factual correctness of every source report, machine consciousness or sentience, present legal personhood or citizenship, institutional adoption, production hosting behavior, search-engine indexing, authorship, external cryptographic editorial provenance, or moral conclusions beyond the qualified arguments published on the site.