1. Define the shared AI and IP perimeter
Identify products, models, code, data, patents, content, know-how and people used by both businesses.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an AI-and-IP perimeter charter.
The principal failure occurs when legal-entity ownership describes operational dependency. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for define the shared ai and ip perimeter should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
2. Map business-critical use cases
Connect each AI capability and IP asset to products, revenue, customers, controls and Day-One processes.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a use-case dependency map.
The principal failure occurs when an asset list proves business criticality. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map business-critical use cases should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
3. Build the chain of title
Trace employee, contractor, founder, university, joint-development and acquired rights.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a chain-of-title dossier.
The principal failure occurs when payment proves assignment. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for build the chain of title should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
4. Classify the asset stack
Separate patents, copyright, database rights, trade secrets, trademarks, data, models and contractual rights.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a layered asset register.
The principal failure occurs when AI is treated as one indivisible asset. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for classify the asset stack should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
5. Distinguish background and foreground IP
Allocate pre-existing technology, transaction-developed improvements and jointly created results.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a background-foreground matrix.
The principal failure occurs when all improvements follow the transferred business. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for distinguish background and foreground ip should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
6. Trace third-party components
Inventory open-source software, pretrained models, APIs, datasets, libraries and embedded services.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a third-party dependency register.
The principal failure occurs when commercial availability creates transferability. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for trace third-party components should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
7. Map licence rights
Test assignment, sublicensing, change of control, field, territory, duration, exclusivity and termination.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a licence-rights matrix.
The principal failure occurs when a right to use is a right to transfer. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map licence rights should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
8. Map training-data rights
Trace provenance, collection terms, reservations, licences, privacy duties and permitted model uses.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a training-data provenance file.
The principal failure occurs when lawful possession settles training rights. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map training-data rights should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
Table 1. Rights-and-dependency register
| Layer | Core question | Evidence |
|---|---|---|
| model | ownership and permitted use | model file |
| data | provenance and purpose | rights ledger |
| code | authors and licences | software bill |
| know-how | custody and transfer | expert map |
Illustrative analytical design; company-specific evidence and professional advice govern.
Values are illustrative evidence indices and require company-specific support.
9. Map model provenance
Document base models, fine-tuning, retrieval sources, prompts, evaluations and modifications.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a model lineage graph.
The principal failure occurs when model weights reveal the whole lineage. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map model provenance should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
10. Map output rights
Assess human contribution, contractual allocation, confidentiality and downstream restrictions.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an output-rights protocol.
The principal failure occurs when every AI output receives copyright automatically. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map output rights should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
11. Test patent position
Map claims, inventors, owners, jurisdictions, prosecution status and technical contribution.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a patent position schedule.
The principal failure occurs when an AI label establishes patentability. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for test patent position should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
12. Test trade-secret continuity
Identify secrecy measures, access, repositories, disclosure history and residual know-how.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a trade-secret continuity plan.
The principal failure occurs when confidentiality clauses preserve secrecy by themselves. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for test trade-secret continuity should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
13. Separate trademarks and digital identity
Allocate names, domains, model names, handles, app-store assets and customer-facing identifiers.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a digital identity schedule.
The principal failure occurs when corporate-name transfer covers every digital identifier. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for separate trademarks and digital identity should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
14. Map regulatory roles
Identify provider, deployer, importer, distributor, modifier and downstream obligations.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an AI regulatory responsibility map.
The principal failure occurs when the sale agreement determines statutory roles. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map regulatory roles should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
15. Map model documentation
Preserve intended purpose, capabilities, limitations, data summary, testing, risk and energy information.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a model documentation pack.
The principal failure occurs when technical documentation can be recreated after completion. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map model documentation should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
16. Map security obligations
Trace model access, secrets, credentials, supply chain, incident history and adversarial testing.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an AI security control register.
The principal failure occurs when code transfer preserves model security. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for map security obligations should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
Table 2. Separation alternatives
| Alternative | Best fit | Critical test |
|---|---|---|
| transfer | discrete ownership | complete chain of title |
| exclusive licence | retained legal title | economic equivalence |
| shared licence | continuing dual use | governable boundaries |
| rebuild | non-transferable estate | clean capability proof |
Illustrative analytical design; company-specific evidence and professional advice govern.
Values are illustrative evidence indices and require company-specific support.
17. Quantify asset dependency
Measure revenue, margin, renewal, service, product and compliance reliance on each shared asset.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an asset-dependency model.
The principal failure occurs when dependency is adequately described as high or low. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for quantify asset dependency should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
18. Identify indivisible assets
Find models, platforms, datasets, patents and teams whose separation destroys utility.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an indivisibility decision log.
The principal failure occurs when every asset can be divided by contract. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for identify indivisible assets should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
19. Define strategic alternatives
Compare transfer, exclusive licence, non-exclusive licence, field split, clone, rebuild, service and retirement.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an alternatives architecture.
The principal failure occurs when ownership transfer is always superior to licensing. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for define strategic alternatives should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
20. Design a transfer
Specify assets, rights, deliverables, registrations, consents, warranties and operational handover.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a transfer implementation plan.
The principal failure occurs when a schedule of titles completes an AI transfer. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for design a transfer should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
21. Design an exclusive licence
Define exclusivity, field, territory, duration, improvements, enforcement and reversion.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an exclusive-licence term sheet.
The principal failure occurs when exclusive use is economically identical to ownership. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for design an exclusive licence should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
22. Design a non-exclusive licence
Allocate shared use, competition boundaries, service levels, information rights and conflicts.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a shared-use governance plan.
The principal failure occurs when shared rights create no operating friction. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for design a non-exclusive licence should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
23. Design a field-of-use split
Separate customers, products, geographies, industries and applications with measurable boundaries.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a field-of-use control map.
The principal failure occurs when field definitions remain stable as models evolve. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for design a field-of-use split should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
24. Design a model clone
Test whether weights, data rights, pipelines, tooling, documentation and expertise can be duplicated.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a clone feasibility assessment.
The principal failure occurs when copying a repository creates a portable model. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for design a model clone should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
Table 3. Model portability stack
| Component | Separation dependency | Proof |
|---|---|---|
| weights | rights and compatibility | controlled copy |
| training estate | provenance and access | lineage record |
| pipeline | code and infrastructure | rebuild test |
| operations | monitoring and expertise | service rehearsal |
Illustrative analytical design; company-specific evidence and professional advice govern.
Values are illustrative evidence indices and require company-specific support.
25. Design a rebuild
Define functional requirements, clean-room inputs, people, data, validation, cost and timetable.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a clean rebuild plan.
The principal failure occurs when rebuild cost equals engineering payroll. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for design a rebuild should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
26. Design transitional access
Time-bound APIs, repositories, experts, tooling, hosting and support.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an AI transition-service schedule.
The principal failure occurs when temporary access can remain informal. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for design transitional access should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
27. Allocate improvements
Determine ownership and use of fine-tuning, adaptations, feedback, derivative models and inventions.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an improvement-rights mechanism.
The principal failure occurs when improvements belong automatically to the operator. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for allocate improvements should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
28. Allocate enforcement
Assign monitoring, prosecution, defence, settlement, cooperation, cost and recovery.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an enforcement protocol.
The principal failure occurs when the legal owner will enforce for both parties. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for allocate enforcement should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
29. Control open-source obligations
Trace licences, notices, source obligations, attribution, copyleft and model terms.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an open-source compliance bill.
The principal failure occurs when open source means unrestricted commercial transfer. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for control open-source obligations should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
30. Protect competition-sensitive information
Design clean teams, access purpose, output controls and expiry for shared technology.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a technology clean-team protocol.
The principal failure occurs when an NDA resolves pre-close competition risk. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for protect competition-sensitive information should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
31. Plan employee and know-how transfer
Map inventors, maintainers, domain experts, restrictions, incentives and documentation.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a critical-knowledge transfer plan.
The principal failure occurs when code and patents substitute for tacit knowledge. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for plan employee and know-how transfer should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
32. Define valuation scenarios
Value transfer, licence, rebuild, delay, degradation and stranded assets through cash-flow effects.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an AI-and-IP value bridge.
The principal failure occurs when book value captures AI and IP economics. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for define valuation scenarios should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
Table 4. Value-and-risk bridge
| Driver | Value channel | Control |
|---|---|---|
| licence loss | revenue and margin | consent gate |
| rebuild delay | cash flow and launch | milestone plan |
| model decay | quality and renewal | validation suite |
| claim exposure | cost and injunction | protection matrix |
Illustrative analytical design; company-specific evidence and professional advice govern.
Values are illustrative evidence indices and require company-specific support.
33. Price execution risk
Quantify consent failure, licence loss, claim exposure, rebuild delay, model decay and service disruption.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a risk-adjusted consideration bridge.
The principal failure occurs when legal risk belongs outside valuation. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for price execution risk should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
34. Draft transaction protections
Link title, provenance, compliance and operability evidence to conditions, covenants, indemnities and price.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an evidence-linked protection matrix.
The principal failure occurs when broad IP warranties cure missing diligence. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for draft transaction protections should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
35. Test Day-One operability
Rehearse access, inference, integrations, monitoring, rights, expertise and incident response.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a Day-One AI operability certificate.
The principal failure occurs when legal completion proves technical operability. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for test day-one operability should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
36. Test seller continuity
Confirm retained rights, tooling, data, people, brands and controls remain usable.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a retained-business continuity certificate.
The principal failure occurs when buyer operability is the only separation test. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for test seller continuity should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
37. Govern the interim state
Control change, retraining, releases, filings, incidents, access and improvements before separation closes.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an interim AI governance register.
The principal failure occurs when a general conduct covenant controls model evolution. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for govern the interim state should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
38. Execute registrations and notices
Complete assignments, licence records, domain changes, consents and counterparty notices.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is a perfection and notice checklist.
The principal failure occurs when contract signature updates every external register. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for execute registrations and notices should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
39. Certify independent capability
Evidence lawful rights, technical portability, operational control, security and accountable ownership.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an independent-capability certificate.
The principal failure occurs when successful inference proves independence. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for certify independent capability should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
40. Close residual entanglement
Resolve shared accounts, copies, model lineage, licences, repositories, claims, support and future improvements.
The controlled record should identify the relevant customer, revenue stream, contract, cohort, relationship owner, commercial dependency, evidence source, baseline, trend, scenario, control, exception, accountable executive, deadline and approval. The immediate deliverable is an AI-and-IP separation closure report.
The principal failure occurs when project closure eliminates residual AI and IP dependency. Reviewers should connect the evidence to revenue durability, gross margin, cash conversion, renewal probability, switching behaviour, counterparty credit and transaction value; distinguish contractual protection from observed customer behaviour; and test whether the conclusion survives a downside scenario.
The decision pack for close residual entanglement should state the commercial question, measurement perimeter, historical evidence, customer-specific facts, forecast logic, sensitivity, management action, buyer implication, residual uncertainty and next gate. Material exceptions should flow into the valuation model, quality-of-earnings work, diligence room, sale-process narrative, transaction protections and board reporting.
Table 5. Closure certificate
| Domain | Required evidence | Accountable owner |
|---|---|---|
| rights | title, licence and filings | legal |
| technology | portable working capability | technology |
| operations | people, process and service | business |
| risk | security and compliance | risk executive |
Illustrative analytical design; company-specific evidence and professional advice govern.
Values are illustrative evidence indices and require company-specific support.
References
- World Intellectual Property Organization, WIPO and Artificial Intelligence, https://www.wipo.int/en/web/wipo-and-ai
- World Intellectual Property Organization, Frequently Asked Questions: AI and IP Policy, https://www.wipo.int/en/web/frontier-technologies/artificial-intelligence/faq
- World Intellectual Property Organization, The WIPO Conversation on Intellectual Property and Frontier Technologies, https://www.wipo.int/en/web/frontier-technologies/frontier_conversation
- World Intellectual Property Organization, IP Policies for Universities and Research Institutions, https://www.wipo.int/en/web/technology-transfer/ip-policies
- European Commission, AI Act, https://digital-strategy.ec.europa.eu/en/factpages/ai-act
- European Commission, Guidelines on the Scope of Obligations for Providers of General-Purpose AI Models, https://digital-strategy.ec.europa.eu/en/library/guidelines-scope-obligations-providers-general-purpose-ai-models-under-ai-act
- European Commission, General-Purpose AI Code of Practice, https://digital-strategy.ec.europa.eu/en/policies/contents-code-gpai
- European Commission, Guidelines for Providers of General-Purpose AI Models, https://digital-strategy.ec.europa.eu/en/policies/guidelines-gpai-providers
- European Commission, Standard Contractual Clauses for International Transfers, https://commission.europa.eu/law/law-topic/data-protection/international-dimension-data-protection/standard-contractual-clauses-scc_en
- UK Government, Copyright and Artificial Intelligence Consultation, https://www.gov.uk/government/consultations/copyright-and-artificial-intelligence/copyright-and-artificial-intelligence
- UK Intellectual Property Office, Examining Patent Applications Relating to Artificial Intelligence Inventions, https://www.gov.uk/government/publications/examining-patent-applications-relating-to-artificial-intelligence-ai-inventions
- UK Intellectual Property Office, Examining Patent Applications Involving Artificial Neural Networks, https://www.gov.uk/government/publications/examining-patent-applications-involving-artificial-neural-networks
- US Copyright Office, Copyright and Artificial Intelligence, https://www.copyright.gov/ai/
- US Patent and Trademark Office, Revised Inventorship Guidance for AI-Assisted Inventions, https://www.uspto.gov/subscription-center/2025/revised-inventorship-guidance-ai-assisted-inventions
- US Patent and Trademark Office, FAQs on Inventorship Guidance for AI-Assisted Inventions, https://www.uspto.gov/initiatives/artificial-intelligence/faqs
- European Patent Office, Guidelines for Examination: Artificial Intelligence and Machine Learning, https://www.epo.org/en/legal/guidelines-epc/2025/g_ii_3_3_1.html
- NIST, Artificial Intelligence Risk Management Framework, https://www.nist.gov/itl/ai-risk-management-framework
- NIST, Cybersecurity Framework 2.0, https://www.nist.gov/cyberframework
- NIST, Secure Software Development Framework, https://csrc.nist.gov/projects/ssdf
- NIST, Cybersecurity Supply Chain Risk Management Practices, https://csrc.nist.gov/pubs/sp/800/161/r1/upd1/final
- OECD, AI Principles, https://oecd.ai/en/ai-principles
- Open Source Initiative, Open Source AI Definition, https://opensource.org/ai/open-source-ai-definition
- SPDX, Software Package Data Exchange Specification, https://spdx.dev/specifications/
- International Organization for Standardization, ISO/IEC 42001 Artificial Intelligence Management Systems, https://www.iso.org/standard/81230.html
- International Organization for Standardization, ISO/IEC 27001 Information Security Management, https://www.iso.org/standard/27001
- International Valuation Standards Council, International Valuation Standards, https://ivsc.org/standards/
