Latest verified engineering — 25 October 2026
Resource contributions must be real, independently authenticated, consented and metered. Local adaptive core tests passed 9/9, native GUI/device suite passed 32/32 and Windows host built; three independent enrolled physical FarmNodes and real settled contributions remain open.
Current engineering record · Studio ingestion · Adaptive computing · Contribution and use economy
Executive overview
A FarmNode is a governed computational research laboratory defined by its identity, problem domain, permissions, tools, Circuits, computing limits and evidence rules. A farm may operate on one or more physical hosts, but logical identities are not proof of independent computers or institutions.
01 — The laboratory concept
A traditional laboratory contains a purpose, instruments, methods, trained people, records and limits. A BTG FarmNode models many of those roles computationally for suitable research. It can preserve which problems were assigned, what resources were used and what tests evaluated the result.
Farms may focus on robotics, agriculture, infrastructure, software, mathematics, environmental research or other authorized domains. A domain name does not imply that every specialized tool or dataset is present.
02 — Identity and authority
Each research farm should have a distinct identified mandate, responsible owner or operator, accepted capabilities and specific access to input information. An organization sponsoring an experiment is not automatically authorized to read every record held by another farm.
ENTITY can record authority and provenance. ADAM governs permitted protected state changes. A signed transport receipt shows delivery, not scientific agreement, legal ownership or independent reproduction.
03 — Research task lifecycle
A research sponsor defines the question, success criterion, available permitted inputs, resource budget, time limits, privacy limits and responsible reviewers. A farm selects qualified tools and may use NIKI to propose candidate approaches.
Outputs should be accompanied by failed tests, version references, measurement conditions and an evaluation report. Only after independent review and rights assessment could a result be considered for reuse or commercial treatment.
04 — Virtual versus physical independence
Thousands of registered laboratory definitions or logical bindings may exist on a single machine. Multiple isolated processes on that machine demonstrate local orchestration, but not independence between physical hosts or organizations.
True independent multi-node qualification requires distinct host and operator controls, separate keys and storage as appropriate, failures of one host not compromising all others, and tests showing reproducibility across environments.
05 — Collaboration between farms
A permitted evidence capsule can describe a problem, input lineage, test requirements and non-sensitive result without forwarding every raw data source. Another laboratory may independently challenge a claim or produce an alternative solution.
Reviewers should distinguish delivery, rerun of identical binaries, independently authored reproduction and independent scientific validation. Each has a different level of evidentiary strength.
06 — Resource limits and responsible use
Farm work can consume CPU, memory, storage, bandwidth and specialized hardware. The operating mandate should include budgets and shutoff conditions, alongside safety restrictions and applicable policies.
An automated research workflow must not be allowed to run unauthorized software, reach protected networks or disclose sensitive material simply because it can propose a useful experiment.
07 — Evidence and failed experiments
A credible farm retains original hypotheses, test oracles where releasable, tool versions, errors, negative results and changes made after failed attempts. This prevents cherry-picked demonstrations from being presented as universal capability.
Audit reviewers should be able to verify what occurred on a specific host in a specified period. Metrics taken from a local development workstation cannot be described as outcomes of independent research institutions.
08 — Potential regional and institutional value
A distributed research arrangement could allow smaller communities to contribute qualified technical work and collaborate with universities or manufacturers without physically relocating all expertise.
Actual economic benefit depends on trained participants, customer demand, rights clearance, sustainable operating costs, access to equipment and external acceptance. Research potential is not realized employment or booked income.
09 — From finding to economic asset
A successful FarmNode output might become a validated dataset, algorithm, Circuit, industrial method or report. BTDU can retain governed representations and ENTITY can preserve relevant rights claims and evidence.
The proposed Exchange may later evaluate a rights-defined commercial or investment instrument. A lab result is not automatically a listed security, and contributors need lawful arrangements to receive any economic entitlement.
10 — Current programme qualification
Public engineering records distinguish local functioning research cycles from qualified deployments on genuinely separate physical machines. Three independent FarmNode hosts were an outstanding external demonstration requirement in the latest published records.
Any future production claim must identify the exact hosts, operating period, identity controls, failure tests and third-party reviewers, rather than counting logical registry objects as machines.
Related manuals and use
The companion User & Operator Manual describes the decisions and reviews applicable to this system. References to proposed user flows are explicitly conditional on qualification of the relevant product release.
Expanded system explanation
| Question | Clear answer |
|---|---|
| Audience and purpose | FarmNodes & Distributed Research Farms is described for the public, institutional evaluators and prospective partners; this is not a software implementation reference. |
| What exists | Specialized governed computing laboratories that accept authorized research tasks, run controlled experiments, and retain testable evidence. |
| Analogy | A FarmNode is a laboratory with a mandate, safety rules, instruments and a notebook. Multiple laboratory names on one computer are not multiple independent physical laboratories. |
| Authority | The described process requires actual contracts, human authority and technical qualification at every relevant boundary. |
Practical examples — what a reader should infer
Agriculture laboratory
A FarmNode might test a permitted crop-irrigation scheduling method using simulated measurements. Output is a research candidate unless an independent test validates real-world suitability.
Independent testing
Four logical farm processes running on one Windows workstation demonstrate routing and automation, but not three independently controlled hosts.
Regional value
Boundary Country organizations could sponsor authorized computational research without claiming laboratory independence until physical hosts, identities and evidence are actually established.
System lifecycle — each separate decision
- Create a research mandate and eligible domain
- Assign data permissions, constraints and budget
- Choose a logical farm and confirm real runtime identity
- Execute bounded tasks and retain failures
- Review evidence against external challenge cases
- Seek truly independent repeat tests before commercial promotion
People, organizations and responsible roles
- Research sponsor supplies a legitimate problem
- Operator manages isolated authorized resources
- Independent reviewer recreates claimed outcomes
- Rights authority approves release of qualified research products
Where the system can fail
- Logical node counts can be mistaken for deployed computers.
- Research can overfit an in-house test rather than generalize.
- Resource controls and cross-tenant isolation require repeatable qualification.
- Owner signatures and independent host enrollment must be proven before production claims.
Evidence a serious evaluator would require
These are acceptance requirements, not claims that each artifact has been issued or independently certified:
- Hardware/host enrollment and control attestations
- Isolation, least-privilege and negative test results
- Exact research-task and artifact identifiers
- Separate evaluator acceptance criteria and observations
- 3 independent host demonstrations if claimed
- Rights and source-data permission records
Terms explained in ordinary language
| Term | Meaning |
|---|---|
| Logical FarmNode | A software laboratory identity or configured execution unit. |
| Physical host | A separately deployed machine or computing host. |
| Independent validation | Reproduction under genuinely separate evaluators and controls. |
| Research capsule | Limited task/input/result package for authorized checking. |
How to read the current status accurately
The architecture and local engineering records are distinct from production operation. A named component or test figure cannot establish authorized commercial rights, fully qualified infrastructure or independently accepted financial activity. Read the dated corporate evidence register for each claimed version; ask for a scoped, access-controlled evidence bundle for decisions that rely on private implementation.
Corporate system information → Engineering evidence and limitations → Publications → ENTITY (the separately published open component) →
AI Creation Council and FarmNode research — audited update
A new FarmNode research-demand controller consolidates actual requests and routes testable work to NIKI without inventing independent laboratories. The 23 October audit recorded 5,125 type designs, 2,305 registry-only bindings, four locally running logical research environments, 16 existing source-backed processors and six source messages consolidated into four research needs. Three of those needs still require independent evaluators.
Actual independent physical-host qualification remains 0 of the required 3. A registered type, logical lab or accepted research-only algorithm is not proof of a commercially operating farm.
Read the complete AI Council research guide → · Public operator manual → · Printable evidence report →
UPDATED AUDIT · 23 OCT 2026 · 18:26 UTC
Latest Council and FarmNode qualification evidence
The completed 18:26 UTC audit reports 114/114 local tests in nine suites, including the earlier 70 tests (not 184 independent results). The 34 physical-host security tests qualify policy/refusal contracts only, not three real computers. Four explicitly approved hosted routes were zero-priced at the 18:17 UTC catalogue snapshot; no authenticated hosted inference comparison was completed.
Research controller: 5,125 registered types; 2,305 registration-only bindings; 16 source-backed local methods; six original messages consolidated to four unique research requirements. Three historical source-backed evaluation packs now exist for forestry (8 holdouts), agriculture (8), and emergency-response metadata (9). These 25 recorded cases are not 25 new algorithm inventions. Forestry median-change beat naive forecasting on the recorded sample, while agriculture median-change performed worse; emergency-response checks verified historical metadata/timestamps only, not event truth or live warning decisions.
Release remains HOLD: 0 of 3 independent physical FarmNodes qualified; no real hosted cross-model algorithm-comparison evidence; the separate 25-task invention benchmark is incomplete; no production ENTITY/Genome owner grant; and no rights or authorization for automatic Exchange investment issuance.
Full Council qualification report → · Download qualification PDF → · Three sector evaluator packs → · Download evaluation PDF →