artifacts/standard-named

Rendered from markdown source. Open raw source on GitHub.

Semantic Integrity Initial Due Diligence FAQ

Purpose: This FAQ is designed for early conversations with a potential design partner, first client, or strategic investor. It separates questions about the pilot from questions about the investment opportunity. It is intentionally non-promotional and avoids pricing, valuation, or deal terms.

---

Section 1: Pilot FAQ

What is the purpose of the pilot?

The purpose of the pilot is to test whether Semantic Integrity can make a real regulated workflow more legible, auditable, and safely assistable by private or local AI systems.

The pilot is not intended to transform the entire organization. It focuses on one bounded workflow where policies, roles, evidence, decisions, approvals, exceptions, and handoffs can be mapped clearly enough to test whether structured semantic context improves AI usefulness and human oversight.

What is Semantic Integrity trying to prove during the pilot?

The pilot is meant to test whether structured semantic containers can improve the performance and governability of AI in regulated organizational work.

The central proof is not that AI can generate text. The proof is whether local or sovereign AI becomes more useful when it is given explicit, human-readable context about the organization’s policies, roles, authority boundaries, evidence requirements, and review obligations.

Why does Semantic Integrity need a real client environment to prove this?

The most important claims depend on real organizational complexity. Synthetic workflows can demonstrate the method, but they cannot fully prove the operating value.

A real client environment provides the actual policies, procedures, documents, review habits, exceptions, ambiguity, security constraints, and role boundaries that determine whether the system is useful in practice.

What kind of workflow is appropriate for an initial pilot?

The best initial workflow is important enough to matter but narrow enough to map within a bounded period.

Good candidates include client onboarding, month-end review, compliance exception routing, document intake, regulated approval workflows, vendor review, audit preparation, or any process where human judgment, evidence, authority, and repeatable procedure intersect.

The initial workflow should involve real documents, real handoffs, and real review obligations, but it should not be the most mission-critical or fragile workflow in the organization.

What does Semantic Integrity produce during the pilot?

Typical pilot outputs include:

• A workflow map
• A role and authority map
• An evidence and source map
• A policy and procedure conversion
• A human review and escalation model
• A semantic container representation of the workflow
• A local or private AI test harness, where appropriate
• A comparison of AI performance with and without structured semantic context
• An auditability and traceability review
• A final pilot report summarizing what worked, what failed, and what should happen next

What is a “semantic container” in practical terms?

A semantic container is a human-readable structure that defines the meaning, boundaries, inputs, outputs, authority, dependencies, and evidence requirements of a workflow or workflow component.

In practice, this may include plain-English documentation, structured prompts, YAML, JSON, checklists, workflow descriptions, role definitions, policy references, approval rules, and audit trail logic.

The goal is not to force people to adopt new jargon. The goal is to make existing organizational meaning explicit enough for both humans and AI systems to work with it safely.

Is this a software platform?

Not in the conventional sense. Semantic Integrity should not be understood primarily as a closed software platform that traps workflows inside a proprietary system.

The approach is based on making organizational work portable, human-readable, and structurally clear. Tooling may exist or emerge around visualization, conversion, audit, and insight, but the core value is the method of converting organizational meaning into governed, reusable semantic structures.

Does the client become locked in?

No. Portability is a feature of the approach.

The client should be able to retain and understand the workflow artifacts created during the pilot. The goal is not to create dependence through obscurity. The goal is to create value through clarity, structure, auditability, and continued improvement.

If the client can leave, why would they continue working with Semantic Integrity?

The ongoing value is expected to come from maintenance, refinement, domain expertise, conversion tooling, local/private AI performance techniques, audit and visualization tools, and the ability to keep semantic structures current as the organization changes.

The relationship should be retained because it is useful, not because the client is trapped.

What is meant by “private,” “local,” or “sovereign” AI?

These terms refer to AI deployments that avoid unnecessary exposure of sensitive data to uncontrolled external systems.

Depending on the client’s needs, this may mean local models, on-premise deployment, private cloud environments, client-controlled infrastructure, restricted inference pathways, or other architectures designed to preserve confidentiality and data control.

The exact deployment model depends on the client’s regulatory obligations, technical environment, and risk tolerance.

Why not just use a powerful cloud AI model?

For many regulated workflows, the issue is not simply model quality. It is data boundary control, auditability, authority, reviewability, and client trust.

Cloud AI may be useful for some tasks, but sensitive workflows often require clearer guarantees about where data goes, who controls the runtime, what the model is allowed to do, and how outputs are reviewed.

Semantic Integrity focuses on making AI assistance compatible with organizational authority and evidence requirements.

What does “performative local AI” mean?

Performative local AI means that private or local AI systems become useful enough to perform real work within bounded, regulated workflows.

The claim is not merely that local models can run. The claim is that structured semantic context, workflow boundaries, and auditable context injection can make local or sovereign AI meaningfully more useful than it would be with generic prompting alone.

How is the pilot evaluated?

The pilot should be evaluated against practical criteria, such as:

• Does the mapped workflow become clearer to humans?
• Does the AI perform better with Semantic Integrity context than without it?
• Does the system reduce review burden or ambiguity?
• Are outputs easier to audit?
• Are authority boundaries preserved?
• Are unsupported claims, hallucinations, or inappropriate actions reduced?
• Can the workflow artifacts be exported, reviewed, and maintained?
• Does the client see a credible path from pilot to production?

What comparison should be used during the pilot?

A useful pilot compares several modes:

1. The current human-only workflow
2. AI assistance without Semantic Integrity structure
3. Local or private AI without structured semantic context
4. Local or private AI with Semantic Integrity context
5. Human-reviewed AI outputs with audit trails

This comparison helps determine whether the semantic layer creates measurable value.

Does the system make decisions automatically?

The initial focus should be assistance, review, routing, explanation, preparation, and audit support rather than autonomous decision-making.

In regulated contexts, Semantic Integrity should preserve the distinction between execution and authority. AI may assist, draft, summarize, flag, classify, or recommend, but human authority remains explicit wherever the workflow requires it.

How are human review and approval handled?

Human review is treated as part of the workflow, not an afterthought.

The pilot should identify which decisions require human judgment, who has authority, what evidence they need, what exceptions trigger escalation, and how approvals or overrides are recorded.

What happens when the AI is uncertain or encounters an exception?

The system should route uncertainty and exceptions back to human review.

A successful pilot should show that the workflow does not fail silently when ambiguity arises. Instead, the system should identify uncertainty, preserve relevant context, and escalate appropriately.

What is the audit trail supposed to show?

An effective audit trail should make it possible to trace an output or recommendation back to the relevant source materials, policies, role definitions, authority boundaries, evidence requirements, and human approvals.

The audit trail should help answer: why was this output produced, what information was used, what rules or context applied, and who approved or changed the result?

What does the client retain after the pilot?

The client should retain the workflow artifacts specific to its own operations, including mapped procedures, role definitions, evidence structures, and other agreed deliverables.

The client should be able to understand and use those artifacts independent of any single interface.

What does Semantic Integrity retain after the pilot?

Semantic Integrity should retain its general methods, reusable tooling, templates, implementation techniques, non-client-specific learnings, and any underlying IP not specifically assigned or licensed to the client.

Client confidential information, proprietary procedures, and sensitive data should remain protected according to the engagement terms.

What are the main risks of the pilot?

The main pilot risks are:

• The selected workflow may be too broad or poorly bounded
• Local AI may not perform well enough for the chosen task
• The semantic mapping may take more effort than expected
• The review burden may not decrease
• The client’s existing procedures may be too inconsistent to encode cleanly
• The pilot may reveal that the organization needs process clarification before AI assistance can be valuable

These risks are useful to discover early.

What would count as a successful pilot?

A successful pilot does not require full production deployment.

A successful pilot demonstrates that one real workflow can be mapped into a human-readable semantic structure, that AI assistance improves when structured context is applied, that human authority and review remain clear, and that the resulting artifacts are auditable and portable.

What would count as an unsuccessful pilot?

An unsuccessful pilot would show that the method does not materially improve AI usefulness, does not reduce ambiguity, does not improve auditability, or requires so much manual maintenance that it is not commercially practical.

An honest negative result is still valuable because it clarifies where the approach needs refinement.

---

Section 2: Investor FAQ

What is the investment thesis?

The investment thesis is that regulated organizations will need infrastructure and implementation methods that allow them to use AI without losing control of data, authority, evidence, auditability, or institutional meaning.

Semantic Integrity aims to occupy the space between governance, workflow engineering, private AI deployment, and organizational knowledge preservation.

What is the core moat?

The core moat is not the fact that workflows can be written in YAML, JSON, prompts, or plain English.

The core moat is whether Semantic Integrity can make private or local AI perform better inside real workflows by transforming organizational rules, roles, evidence, and authority boundaries into usable context structures, context injections, tuned model behavior, or other auditable execution patterns.

If that claim is proven, the defensibility comes from method, accumulated implementation knowledge, tooling, domain-specific templates, conversion processes, and performance techniques for sovereign AI environments.

Is there a traditional platform moat?

Not at the current stage.

Semantic Integrity should not be evaluated as a conventional locked SaaS platform. The current approach intentionally emphasizes portability and client control. This reduces traditional lock-in, but it also aligns with the trust needs of regulated clients.

Future proprietary tooling may create additional product defensibility, especially around visualization, audit, insight, conversion, maintenance, and performance optimization.

Is portability a weakness or a strength?

It is both.

For clients, portability is a strength because it reduces adoption risk. They are not trapped if the company changes direction or ceases support.

For investors, portability weakens the usual lock-in model. The investment case must therefore rely on demonstrated value, implementation depth, performance improvement, tooling, maintenance, and trusted expertise rather than captivity.

Why would customers stay if they can leave?

Customers would stay if Semantic Integrity continues to provide value that is difficult to reproduce internally.

That value may include better local AI performance, faster workflow conversion, ongoing maintenance, regulatory adaptation, audit tooling, visualization, insight generation, model adaptation, and accumulated domain expertise.

What is the current business model?

The early business model is implementation- and maintenance-centric.

This means revenue may initially resemble a services or consulting model rather than pure software subscription revenue. Over time, tooling, templates, dashboards, audit layers, visualization systems, and vertical-specific packages may create more scalable product revenue.

Is this a services business or a product business?

At the earliest stage, it is likely a services-led product company.

The service work is necessary to discover real workflows, build proof, validate patterns, and create reusable structures. The product opportunity emerges as repeated implementation patterns become tools, templates, dashboards, and licensed systems.

What should an investor be most skeptical about?

An investor should be skeptical about whether the local/private AI performance advantage is real, measurable, and repeatable.

The strongest diligence question is: does Semantic Integrity produce better results than a competent team using ordinary SOPs, generic prompts, and available AI tools?

Other areas for skepticism include scalability, founder dependency, maintenance burden, sales complexity, terminology friction, and the transition from services to product.

What evidence would increase investor confidence?

Investor confidence should increase if the company can show:

• A real workflow converted into semantic containers
• Improved AI performance with structured context
• Reduced human review burden
• Stronger auditability
• Clearer authority boundaries
• Exportable and understandable artifacts
• Repeatable conversion patterns
• Evidence that future tooling can reduce implementation labor
• A credible path from first pilot to verticalized offering

What evidence would weaken the investment case?

The investment case weakens if:

• The method does not improve AI outputs
• Local models remain too weak or costly for practical use
• Workflow mapping is too labor-intensive
• Clients do not understand or value the audit/semantic layer
• The system depends entirely on founder judgment
• The approach cannot be repeated across clients
• The proprietary tooling does not materially improve delivery speed or insight quality

What is the role of future proprietary tooling?

Future proprietary tooling may include visualization, workflow conversion, audit review, insight dashboards, maintenance systems, semantic diffing, evidence mapping, or model-context optimization tools.

These tools may become important sources of scalability and defensibility. However, they should be evaluated as future upside unless already working.

What is the significance of source escrow or open sharing?

Source escrow or open sharing reduces strategic risk for early partners. It helps reassure clients or investors that they will not be stranded if the company fails, pivots, or loses key personnel.

This may make early adoption easier, especially for regulated organizations. However, it also means the investment thesis should not depend on secrecy alone.

How should IP rights be understood?

There should be a clear distinction between:

• Client-specific workflow artifacts
• Semantic Integrity’s general methods
• Reusable templates and tools
• Proprietary conversion or optimization techniques
• Future software products
• Confidential client data
• Rights for internal use
• Rights for resale, sublicensing, or commercialization

Early discussions should clarify what the client may use internally, what Semantic Integrity retains, and what rights an investor receives.

Does an investor need exclusive rights?

Not necessarily.

Exclusive rights may be appropriate only in narrow circumstances, such as a defined vertical, region, or channel, and only if they create enough value for both sides.

Preferential access, internal-use rights, referral economics, or limited vertical rights may be more practical than broad exclusivity.

What is the risk of giving broad IP access to an early investor?

Overly broad IP grants can create future financing problems, limit commercialization, confuse ownership, or reduce the company’s attractiveness to later investors.

A better structure is usually a durable internal-use license, continuity rights, and clearly defined access to specific tools or artifacts, rather than unrestricted ownership of all present and future IP.

How important is the first client?

The first client is extremely important because the core proof requires contact with real organizational complexity.

The first client functions as a design partner, implementation environment, reference case, evidence source, and stress test for the method.

What makes a good first design partner?

A good first design partner has:

• Regulated or sensitive workflows
• Meaningful AI governance concerns
• Real operational pain
• Access to internal procedures and documents
• Willingness to participate in mapping and review
• Enough authority to run a pilot
• Enough strategic imagination to understand the category
• A practical need for private, local, or controlled AI assistance

What makes a poor first design partner?

A poor first design partner is one that wants broad transformation without operational focus, cannot provide access to real workflow details, lacks internal ownership, expects instant automation, or is unwilling to participate in review and validation.

The first pilot needs a partner who understands that the work is both implementation and proof.

What is the likely path from first pilot to scalable company?

A plausible path is:

1. Complete one narrow regulated workflow pilot
2. Produce evidence of performance, auditability, and review improvement
3. Convert the learnings into reusable templates and tools
4. Repeat in adjacent workflows within the same vertical
5. Package common patterns into vertical-specific offerings
6. Build proprietary tooling around conversion, visualization, audit, and maintenance
7. Expand into additional regulated verticals once repeatability is demonstrated

What are the biggest execution risks?

The biggest execution risks are:

• Insufficient local AI performance
• Too much implementation labor
• Difficulty explaining the category
• Over-customization for the first client
• Lack of repeatability
• Founder dependency
• Slow development of proprietary tooling
• Long enterprise sales cycles
• Confusion between governance documentation and operational AI value

What is the category?

Semantic Integrity sits across several existing categories, including AI governance, workflow engineering, knowledge management, compliance technology, private AI deployment, audit infrastructure, and organizational operating systems.

The category may need to be explained carefully because the company is not simply selling AI automation, compliance software, or documentation management.

What should the initial investor diligence focus on?

Initial diligence should focus on:

• Whether the method improves local/private AI usefulness
• Whether the workflow artifacts are genuinely legible and portable
• Whether audit trails are meaningful
• Whether implementation can become repeatable
• Whether the founder can translate the concept into client-understandable value
• Whether future tooling can reduce labor and improve margins
• Whether early clients will pay for continued maintenance and refinement

What is the cleanest investor framing?

The cleanest framing is that Semantic Integrity is building an operating method and tool layer for regulated organizations that need AI assistance without sacrificing data control, authority, auditability, or institutional meaning.

It should be evaluated as a services-led infrastructure opportunity with potential to become a verticalized product company as repeated patterns become tooling.