Frequently asked questions

At DeepMoat, AI decision intelligence means helping people make consequential choices about AI: what to prove, buy, build, govern, pause, or stop. We start with the business outcome, the people accountable for it, and the consequences of getting the decision wrong. We then compare practical routes against evidence, cost, data boundaries, implementation needs, and operating control. The result is a reasoned next move, which may involve your existing team, another provider, or separately scoped work with DeepMoat.

DeepMoat is for decision makers, institutional leaders, boards, family office managers, senior operators, and entrepreneurs facing a consequential AI commitment involving budget, data, clients, reputation, or how work is done. It is not staff augmentation, generic tool shopping, a free technical review, or a substitute for legal or security specialists. Signal Triage opens the engagement; the next step may be an internal action, another specialist, separately scoped DeepMoat work, or no further engagement. That is DeepMoat’s practical meaning of AI decision intelligence.

Yes. You can bring DeepMoat a business problem, an opportunity, or uncertainty about where AI belongs; you do not need to arrive with a chosen model or a finished specification. Start with what is slow, expensive, unreliable, or difficult to scale, and what a better outcome would look like. In Signal Triage, we explore possible uses, the constraints that matter, and a practical next step. A full workflow assessment or implementation plan requires a separately agreed scope.

Signal Triage is a private, prepared, 45-minute Zoom review of the AI commitment in front of you. At least two core team members join, with advance research into your company and the context you share. You can book and pay immediately. The optional AI intake can help focus the meeting before or after you book; do not submit credentials or sensitive records through it. Bring a high-level summary of the proposal or demo, decision deadline, and known constraints. You leave with the decision clarified, plausible routes compared, material unknowns surfaced, useful ideas, and a recommended next move. A written meeting summary captures the ideas discussed and recommended next steps. The price is US$375 plus applicable tax, credited toward your next accepted paid DeepMoat scope, which can include Signal Briefing. It is not a roadmap, legal opinion, security audit, or implementation plan.

At least two DeepMoat core team members join your 45-minute Zoom, chosen around the question you bring. The core team is Keri Nakamoto, James Griffing, Joseph Rosenbaum, and Kurtis Nakamoto. Before the meeting, we research your company and review the context you share so we can arrive with relevant questions and ideas. For a private organization, we agree what background is appropriate to provide. The aim is a working conversation grounded in your situation.

You receive a written meeting summary capturing the ideas discussed and DeepMoat's recommended next steps. The conversation clarifies the decision, compares plausible routes, and identifies important unknowns. The summary gives you something to return to and discuss with your team. You can use that guidance whether or not you hire DeepMoat again. Technical diligence, a detailed roadmap, a prototype, and implementation are separate deliverables that require their own agreed scope.

No. You can book directly. The optional AI intake asks five focused questions about the situation and turns your answers into an Intake Clarity Brief: your situation, where the complexity appears, what matters to your team, what remains unclear, and what Signal Triage will clarify. We use it to prepare the meeting. Keep it high level; we will agree how to handle sensitive material before requesting it.

Yes, on either booking path. If your organization has no public website, tick “This organization has no public website” and enter the organization name instead. If you are arranging the session for the decision owner, each path asks about your role. Direct booking asks who you are booking for; choose “The decision owner — I’m authorized to arrange the session”. The optional AI intake asks about your decision involvement; choose “I’m exploring this for the decision owner”. If you would rather talk first, email Keri Nakamoto and keep the initial note high level.

Signal Triage costs US$375 plus applicable tax. Payment is nonrefundable except where a refund is required by law, and one reschedule is included. The fee is credited toward your next accepted paid DeepMoat scope, which can include Signal Briefing. A later engagement remains optional and is scoped separately. Review the booking page and checkout terms before completing your reservation; no later purchase is required to use the ideas and recommendations from your session.

We examine whether the promise survives your operating reality: representative work, data and IP terms, privacy and security questions, integration burden, human review, total cost, model or contract changes, support, and exit path. The answer may be proceed, test narrowly, renegotiate, use another route, wait, or stop. Signal Triage does not replace legal, procurement, code, or cybersecurity diligence.

Name the business outcome, accountable owner, baseline and success measure, full cost, usable data, access rules, whose work changes, required training, monitoring and override path, and scale-or-stop criteria. If those answers are missing, approve only a bounded test or readiness work—not a broad rollout.

Buy or configure when the problem is common and speed matters. Build when your own data or workflow creates meaningful advantage and you can operate what you build. Partner when the opportunity is distinctive but internal capacity is limited. Wait when the evidence, controls, or ownership are not ready. Compare time to value, total cost, control, lock-in, maintenance, and reversibility.

Compare the AI-assisted workflow with a clear baseline for the same work. Look at output quality, time saved, error rates, human review, adoption, and the cost of operating the whole process. Include integration, data preparation, training, monitoring, maintenance, and supplier fees rather than counting only the model subscription. A convincing demo is a starting point for testing. Before scaling, establish what improvement matters, who measures it, and what result would justify continuing, changing course, or stopping.

Scale when results repeat on representative work, people adopt the workflow, errors can be detected, controls hold, and the economics survive real volume. Pause or reset when the remaining unknowns can still be tested. Stop when value cannot be shown, supervision erases the savings, risk cannot be bounded, nobody owns the workflow, or supplier dependency becomes unacceptable.

Yes. DeepMoat can help you decide what to keep, improve, connect, or stop within an existing AI workflow. Bring the work people are trying to complete, the tools they use, where results fail, and how outputs are reviewed. Sometimes the useful next step is better configuration, clearer responsibilities, or training with the tools you already have. A new platform or custom build should earn its place through the problem it solves and the cost of operating it.

A family office can examine AI for internal knowledge retrieval, document preparation, recurring reporting, research organization, and administrative workflows. The useful starting point is a specific task, the information it requires, and the person responsible for checking the result. DeepMoat helps compare the potential benefit with confidentiality, access, review, and operating requirements. Keep human approval around consequential outputs and test with an appropriate data boundary before expanding access. Any private-material review is agreed in scope before information is requested.

DeepMoat can help frame and compare AI operating opportunities across portfolio companies. Separate needs that may be shared, such as training or vendor evaluation, from workflows that depend on a particular company's customers, systems, data, and team. Prioritize a concrete business outcome and a responsible owner in each company. Signal Triage can help identify where to start; a multi-company assessment, detailed diligence, or coordinated implementation requires a separately agreed scope. Do not assume that one company's successful pilot will transfer unchanged to another.

Start with a problem that matters to a specific customer or operator, and define the improvement they would recognize. Then compare an existing product, configuration, partnership, or custom build. DeepMoat helps examine whether the advantage comes from your workflow, data, distribution, service, or speed of execution, and what it will take to maintain that advantage. A first test should answer an important uncertainty before you commit a larger budget or team. The right next move may be a smaller experiment or a simpler existing tool.

Signal Triage is the US$375-plus-tax, prepared 45-minute opening session, with at least two core team members and a written meeting summary. Signal Briefing is a deeper engagement for a larger decision, priced at US$10K plus expenses. Briefing includes preparation, a working session with the people accountable for the decision, and a written recap of the recommendation and next steps. Detailed arrangements and any review of private materials or systems are agreed in writing. The Triage fee is credited when Briefing is your next accepted paid scope.

Yes. Capacity & Buildout is DeepMoat's separately scoped route for a prototype, tool, or workflow and the training needed for your team to operate it. Typical projects are US$25K–$75K; larger builds are quoted. The work follows a decision about what is worth building, with scope and price agreed before it begins. The practical questions include who will use it, who reviews its outputs, who maintains it, and what your team needs to run it with control.

Signal Desk provides year-round access for recurring AI decisions from a team that already knows your organization. It is intended for an ongoing need for decision reviews, direct access between reviews, and reserved capacity, subject to qualification. Pricing starts at US$180K per year. Start by considering how often consequential AI questions arise and whether continuity would make decisions better. A single defined question may be served by a smaller engagement. The scope of ongoing work is agreed before a commitment.

Signal Triage is separately paid and scoped around the decision—not a pre-committed build or staffing sale. The recommendation may be to use your existing team or vendor, choose another specialist, narrow the effort, wait, or stop; you can act on it without hiring DeepMoat again. Any later DeepMoat build is optional and separately proposed after the decision.

No. DeepMoat is model agnostic. We work across OpenAI models, Anthropic’s Claude, Google Gemini, other AI providers, and open-source models. We compare the options against your business goals, data requirements, budget, timing, and operating needs. Our OpenAI Select Partner status supports implementation when OpenAI is the right choice; it does not determine the recommendation. The best next step may involve your current provider, a different model, a combination of models, a limited test, or waiting. Relevant partner economics are disclosed before you commit. Any later implementation work is separately scoped.

Start with the minimum context necessary. Do not provide credentials, raw personal data, privileged or client material, or unrestricted system access through an initial intake. Any later review of private materials is agreed separately, including how they will be shared. DeepMoat can surface questions for your legal, security, and privacy specialists; your designated owners retain those sign-offs and operational accountability, including approval, override, monitoring, and shutdown authority.

DeepMoat can show its own operating experience, explain how it evaluates decisions, and make the promised deliverables concrete without naming private clients. Our AI practice began inside ANVL Entertainment, and that build experience is described in our Notes. The homepage shows an illustrative meeting-summary structure, clearly labeled as an example. Memory Lane demonstrates ongoing attention to source-reviewed AI developments; it is not a client-results case study. Studio is a forthcoming product, not evidence of a completed client engagement.

Studio is DeepMoat's forthcoming paid product for entertainment development, planned for studio.deepmoat.ai. It is being developed around script coverage, story bibles, pitch decks, storyboards, and related story-development work. Studio is currently presented here as in development. Availability, product details, and pricing will be announced when it is ready. Its role on this site is to show the direction of DeepMoat's own product work; it is separate from booking a Signal Triage session.