From Decision Support to Immersive Financial Intelligence

The next generation of financial services will not be defined solely by better data, faster transactions, or more sophisticated analytics. It will increasingly be defined by how effectively humans and artificial intelligence can interpret complex information together.

This is where Data2Space can become a powerful foundation for immersive finance: an environment in which financial data, scenarios, risks, market signals, and strategic options are not merely displayed but experienced, explored, challenged, and continuously interpreted.

A particularly promising component of such an environment is the AI advisory board. The concept is straightforward: instead of asking one AI system for a single answer, an organization creates a structured panel of AI personas representing different perspectives, decision principles, and areas of expertise. As Felix Schlenther describes in his AI FIRST article, his advisory board allows him to consult “twelve board members” within minutes and use their different perspectives to “challenge my own thoughts” and uncover perspectives he might not have considered himself.

For immersive finance, this principle can be taken one step further: the AI advisory board becomes spatially and contextually embedded in the financial data itself.

Data2Space as the Environment for the AI Advisory Board

The core idea behind Data2Space is not simply to visualize data in three dimensions. Its strategic value lies in creating an environment in which complex information can be organized into a spatial and interactive context.

An AI advisory board can use this environment as its shared decision space.

Instead of presenting an investment decision as a conventional dashboard containing dozens of charts, Data2Space could construct an immersive financial scenario containing:

• market and macroeconomic indicators,
• portfolio exposures,
• liquidity and cash-flow positions,
• counterparty relationships,
• risk concentrations,
• customer segments,
• regulatory constraints,
• alternative scenarios,
• historical analogues,
• and projected outcomes.

The AI advisory board then operates on top of this common information space.

This creates a fundamental shift: AI does not merely answer questions about financial data; it interprets a shared financial world.

That distinction matters because complex financial decisions rarely fail because information is completely unavailable. They fail because decision-makers interpret the same information differently, overlook assumptions, anchor on a dominant narrative, or fail to identify what information is missing.

Schlenther’s advisory-board methodology explicitly addresses this problem. Before the AI personas give their opinions, the process separates facts from assumptions and makes the decision-maker’s own bias and concerns visible.

Data2Space can turn this methodological principle into an immersive interface.

The AI Advisory Board as a Spatial Decision Architecture

A Data2Space-based financial advisory board could consist of several specialized AI agents, each occupying a clearly defined role.

For example:

The Risk Advisor
Focuses on downside scenarios, concentration risk, liquidity, stress events, and risk-adjusted returns.

The Macro Advisor
Interprets interest rates, inflation, monetary policy, geopolitical developments, and broader economic conditions.

The Portfolio Strategist
Evaluates allocation, diversification, opportunity cost, and long-term portfolio construction.

The Customer Advisor
Represents the perspective of clients, investors, or counterparties and asks whether a proposed strategy creates understandable and defensible value.

The Regulatory Advisor
Tests decisions against compliance, governance, reporting, and regulatory constraints.

The Contrarian Advisor
Actively searches for weaknesses in the dominant investment thesis and attempts to falsify the prevailing narrative.

The CFO Perspective
Evaluates decisions in terms of capital efficiency, cash flow, balance-sheet implications, and strategic flexibility.

The important point is that these personas should not simply generate different writing styles. They should embody different evaluation frameworks.

This corresponds closely to Schlenther’s approach. His personas are defined through core beliefs, typical questions, evaluation logic, permitted biases, revision rules, and explicit “No-Gos.”

In Data2Space, these different perspectives could become spatially visible.

A risk advisor might highlight concentrations in one region of the financial landscape. A macro advisor could draw attention to dependencies between interest rates and asset valuations. A contrarian agent could isolate the assumptions upon which the investment thesis depends.

The result would be an AI-mediated financial landscape rather than a conventional dashboard.

From Dashboard to Dialogue

Traditional financial interfaces generally answer the question:

What is happening?

An immersive AI advisory environment should also answer:

Why is it happening?
What could happen next?
What are we assuming?
What are we missing?
Which argument is strongest?
Under what conditions would our conclusion change?

This is precisely where the advisory-board methodology becomes valuable.

Schlenther deliberately avoids a simple majority vote. His objective is not to determine which persona receives the most votes, but to understand “which argument carries” and under which conditions that argument stops carrying.

That principle is particularly relevant to financial decision-making.

Imagine an investment committee considering a large allocation to a new asset class. In a Data2Space environment, the committee could see the proposed allocation represented spatially alongside historical performance, liquidity, correlations, macroeconomic dependencies, and stress scenarios.

The AI board could then conduct a structured session:

1. Briefing – the investment decision and available information are introduced.
2. Framing – facts, assumptions, objectives, and potential biases are separated.
3. Selection – the system chooses the most relevant advisory perspectives.
4. Information gap analysis – each advisor identifies information required for a robust judgment.
5. Independent first votes – each AI advisor evaluates the identical information package.
6. Cross-critique – advisors challenge the strongest arguments of the others.
7. Synthesis – the system produces recommendations, consensus, dissent, kill criteria, and next steps.

This seven-stage process is directly adapted from the structure described in the source article.

Data2Space adds an important dimension: the arguments can be connected directly to the underlying financial objects and data.

An advisor should not merely say that liquidity risk is increasing. The system could identify the relevant liquidity cluster in the spatial environment and show which positions, counterparties, assumptions, or market conditions contribute to the conclusion.

Why Multiple AI Perspectives Matter in Finance

Financial institutions already use AI for forecasting, fraud detection, portfolio analytics, customer segmentation, and risk management. The AI advisory-board concept addresses a different layer: the quality of strategic judgment.

One AI model can be highly capable and still produce a narrow answer. Multiple specialized perspectives make disagreement explicit.

The source article identifies this as one of the central benefits of the approach: the board exists to provide perspectives that the decision-maker “alone” might not have considered.

In immersive finance, disagreement itself could become an information layer.

Instead of hiding uncertainty behind one aggregate score, Data2Space could represent:

• areas of consensus,
• areas of disagreement,
• assumptions shared by all agents,
• assumptions challenged by only one agent,
• missing information,
• critical dependencies,
• and conditions that would reverse a recommendation.

This could make AI-supported financial decisions substantially more transparent.

The Orchestrator: A Digital Chairperson for Financial Decisions

The most important technical component would be an AI board chair or orchestrator.

According to the source model, the chairperson receives the briefing, sharpens the decision, selects appropriate members, collects missing information, distributes an identical meeting package, gathers the individual votes, coordinates cross-critique, and synthesizes the final decision document.

In Data2Space, this orchestrator could become the user’s primary interface.

A financial executive might simply say:

“Should we increase our exposure to European infrastructure over the next three years?”

The orchestrator would translate that question into a structured decision.

It could then determine that the appropriate advisory board consists of, for example, a macroeconomist, infrastructure investment specialist, risk officer, liquidity specialist, regulatory expert, and contrarian strategist.

Each receives the same underlying financial information.

The system then compares their assessments.

The user does not need to manage six separate AI conversations. The orchestrator manages the board while Data2Space provides the shared environment in which the reasoning can be inspected.

From AI Personas to Financial Digital Twins

The combination becomes even more powerful when the Data2Space environment incorporates financial digital twins.

A portfolio, company, bank, fund, customer segment, or financial market can be represented as a dynamic object with relationships and dependencies.

AI advisors could then interact with these digital representations.

For example, the risk advisor might ask:

What happens to this portfolio if interest rates remain elevated for another 24 months?

The macro advisor might construct a different scenario:

What happens if inflation falls rapidly and central banks begin easing?

The contrarian advisor might ask:

Which assumption in the current valuation creates the greatest vulnerability?

The Data2Space environment could visualize the resulting changes simultaneously.

This transforms scenario analysis from a static report into an interactive conversation with a financial model.

Keeping Human Judgment in the Loop

There is, however, an important principle that should not be lost in the excitement around AI.

The AI advisory board should support judgment, not replace it.

This is also explicit in Schlenther’s model: the advisory board does not make the final decision for the user. Its purpose is to provide structured perspectives from which the human decision-maker forms their own judgment.

That principle is especially important in finance.

A Data2Space implementation should therefore make the distinction between:

Data
What is known and measured.

Assumptions
What the system believes but cannot directly establish.

AI interpretation
What a particular advisory persona concludes from the information.

Recommendation
What the combined advisory process suggests.

Human decision
What the responsible decision-maker ultimately chooses.

Making these layers visible would create a much more trustworthy form of AI-assisted finance.

From Skill to Multi-Agent Financial Infrastructure

The source article describes two practical architectures: a reusable skill containing multiple personas and a multi-agent system in which each board member operates independently.

For immersive finance, both approaches could be useful.

A skill-based advisory board could be deployed quickly for financial analysts, wealth managers, investment committees, or corporate finance teams.

A multi-agent architecture would be more suitable for institutional applications where independence between perspectives is strategically important.

The latter has a particular advantage: each agent can receive the same information independently, reducing the risk that one agent anchors on another agent’s answer. The source specifically identifies this separation as a major advantage of the multi-agent architecture.

For regulated financial environments, this separation could also support stronger auditability.

Every recommendation could retain:

• the information package provided,
• the agent involved,
• its assumptions,
• its reasoning framework,
• its recommendation,
• the objections raised,
• the information that changed its position,
• and the final human decision.

That creates the foundations of an AI decision audit trail.

A New Model for Immersive Finance

The combination of Data2Space and an AI advisory board suggests a broader conceptual model for immersive finance.

Data2Space provides the world.

It organizes financial information into a spatial, relational, and interactive environment.

AI agents provide the perspectives.

They interpret that environment according to distinct principles, objectives, and risk models.

The orchestrator provides the process.

It ensures that the decision follows a structured sequence rather than becoming an unstructured conversation with a chatbot.

The human provides accountability.

The final decision remains with the responsible person or institution.

This can be summarized as:

Financial Data → Data2Space → AI Perspectives → Structured Debate → Human Decision

Such a model moves beyond the idea of “AI as an assistant.” It positions AI as a structured collective intelligence layer surrounding financial decision-making.

The Strategic Opportunity

The most interesting opportunity for Data2Space may therefore not be to build another immersive visualization tool.

It could become the spatial operating environment for AI-assisted financial decision-making.

The advisory board provides the cognitive architecture. Data2Space provides the experiential architecture.

Together, they could allow financial professionals to move through a decision rather than merely read about it.

They could inspect the financial situation, ask questions, introduce scenarios, activate different AI perspectives, observe disagreements, challenge assumptions, and finally make a decision with a much clearer understanding of both the evidence and the uncertainty.

The source article concludes with a simple five-step process: define the optimal board, enrich its personas through research, build the skill or orchestrator, create the individual members, and continuously test and refine the system.

For Data2Space, the next step would be to add a sixth dimension:

Make the board immersive.

Not just What does the AI think?

But:

Where in the financial system does that conclusion come from?

Which data supports it?

Which assumptions drive it?

Which advisor disagrees?

What would have to change for the recommendation to reverse?

That is where an AI advisory board could evolve from a productivity tool into a new interface for immersive finance—and where Data2Space could provide the spatial layer through which financial intelligence becomes not only accessible, but explorable.

Source: Felix Schlenther, “Wie ich meinen KI-Beirat gebaut habe,” AI FIRST Insights, August 2026. The quotations and methodological references above are based on the supplied text; the Data2Space/immersive-finance applications are an analytical extension of those concepts rather than claims made in the original article. Data2Space may build upon, integrate, or further process services, data, models, content, or other offerings provided by third-party service providers only to the extent that such services have been lawfully acquired or licensed from the respective provider and the applicable provider has authorized their use, integration, and further processing for the intended purpose. Data2Space does not acquire or assume any rights beyond those expressly granted by the respective provider and will not use, integrate, make available, or commercially exploit any third-party services or content in a manner that is prohibited by the applicable license, terms of use, or contractual arrangements. Accordingly, the integration and further processing of third-party services by Data2Space is always subject to the existence of the necessary rights, licenses, and permissions granted by the respective provider.