In Part I of this report, we established the reciprocal relationship between AI and intangible assets. We turn now to how AI creates value. AI value creation can be both direct and indirect, and understanding both dimensions is essential to calculating true ROI.  We conclude with key observations from Parts I and II of this report.

The Intersection of Intangible Assets and Artificial Intelligence Value Creation

AI can have an immediate, direct impact on a company’s financial statements through external or internal monetization:

External Monetization. An AI engine might draw upon existing intangible assets to generate new intangible assets—new content, new software, novel protein instructions—that can then be sold or licensed to generate cash flow. The causal pathway is: existing intangible asset input → AI model → new intangible asset output → external sale or license → cash flow.

Internal Monetization. An organization might apply AI to its own activities to generate efficiencies, improve revenue, or enhance products that can then be sold for better margins. Examples include using AI to interrogate historical invoicing data to reduce invoice cycle times, leveraging predictive analytics to optimize resource allocation, or reducing human involvement in claims handling procedures. The causal pathway is: intangible asset input → AI model → operational improvement (expressed as a new intangible asset output) → cost reduction or revenue enhancement → cash flow.

In either case, the economic impact of AI can be immediate (cash flow today) or delayed (future cash flow). Often AI initiatives result in negative cash flows today and (hopefully) positive future cash flows. However, zooming out, investors—who are fundamentally purchasing a claim on a company’s future cash flows—may change the value they assign to the company today based on their assessment of (i) the probability and (ii) the scale of impact the AI investment will have on medium- to long-term profitability.

Indirect Value Creation

Beyond the direct financial impacts that will ultimately be captured in current or future cash flows (and should likewise be captured in traditional project level ROI analysis), there is a second and less widely appreciated way AI drives value creation: it can also impact the value of the intangible assets the company already owns – often creating substantial residual asset appreciation that is essentially invisible in conventional financial evaluation.

For example, a company may own an old dataset that was previously not cost-effective for humans to analyze. A new AI tool may enable the company to make use of that dataset and generate valuable new outputs, not only leading to direct value creation benefits for the company itself but also increasing the value of the dataset to third parties who can do more with that data than the company itself can.

Put another way, the application of AI to existing intangible assets can result in those assets becoming materially more valuable to third parties, above and beyond any value the company can extract from them itself. This is analogous to how the introduction of fracking technology to Eagle Ford Shale in South Texas saw a land rush between 2008–2010, with acreage values skyrocketing from $1,000 to $10,000 per acre. Local landowners might not have been able to extract the oil beneath their land themselves, but an upstream operator could—and that increased the value of the land for the owner, absent any other benefit.

While such indirect increases in intangible asset value may seem uncommon, they implicitly underpin many of the investment cases behind M&A activity. This frequently occurs when a larger entity calculates that (in addition to operating efficiencies and financial engineering) due to its scale it can generate a higher return on the intangible assets of the target than the target itself can. For example, when Meta acquired Instagram for $1B in 2012, Instagram had zero revenue, no subscription model, and no material monetization. Yet Meta paid $1B because it identified that it could scale and utilize Instagram’s intangible assets (it had no physical assets to speak of) more effectively than Instagram could. It was correct: in 2024, Instagram reportedly generated around $66.9 billion in revenue, with estimated value between $400B and $500B.

AI is a trigger event that creates intangible asset appreciation vectors: existing assets may increase or decrease in value due to the introduction of AI. That change in value is off balance sheet and may not be appreciated by market participants—or even the asset owner itself—unless they are attuned to both the existence and potential of that intangible asset.

In summary, when considering the economic outcome of AI initiatives, both direct and indirect value creation must be accounted for. Considering direct value creation only is likely to overlook critical intangible asset appreciation.

Modeling the Economics of AI Return on Investment

Modern AI systems are still very new (ChatGPT was only launched on November 30, 2022) and reliable measures of ROI for AI remain elusive. Responding to pressure to be seen to be “doing something” with AI, many organizations have adopted a “spray and pray” or tech-driven proof-of-concept approach: investing in a miscellany of initiatives in the hope something will deliver a financial windfall. Others have attempted a VC-style portfolio approach, assuming most initiatives won’t work but a few will “return the portfolio.”

Neither approach would be acceptable in the context of a traditional CAPEX investment, such as constructing a new factory. In such a case the CFO and Board would rightly require a rigorous ROI analysis and a defensible business case. The absence of comparable financial discipline in AI investment cannot be justified by scale: global spending on AI has exceeded approximately $500 billion over the past three years. Deploying capital at this magnitude without the rigor of conventional financial analysis approaches the threshold of fiduciary irresponsibility. This highlights the urgent need for robust methodologies that enable strategic decision-makers to evaluate expected returns and risk-reward trade-offs with greater precision.

Some companies are implementing more traditional ROI calculations: “I spent X and got 3X back.” For example, one CFO told us they intend to spend $100M on AI initiatives with the anticipation of receiving $250M in forecasted benefits over 36 months. While such an approach is useful, it has the limitation that costs and income sit almost entirely on the cash-flow side of financial statements.

This ignores the balance sheet side: namely, the assets that are created, degraded, or improved as a function of AI investments. As a result, AI initiatives appear economically similar to one-off expense projects rather than to investments that build durable, compounding asset value. As outlined in Section 2.4 of Part I of this report, under current accounting rules, most intangible asset investment cannot be capitalized. And because investments in AI primarily involve investments in intangible assets—with only limited exceptions for certain modest categories of software development—the vast preponderance of AI R&D is expensed through the P&L. As a result, many assets generated as a function of R&D effort, including software, data, know-how, content, and regulatory approvals, rarely or never appear on the balance sheet.

This presents a major challenge for AI initiatives: the utilization of existing intangible assets, the creation of new ones, and the change in value of utilized intangible assets is essentially not tracked in financial statements. Accordingly, this value is typically absent from traditional ROI calculations which tend to focus on project costs and near-term benefits, not the value of residual (and invisible) assets created and enhanced. To be unable to calculate true ROI on investments of this scale and criticality is deeply concerning.

The Preconditions for AI Value Creation

A critical caveat: AI is not inherently valuable. Its value depends on preconditions that organizations must actively create and maintain:

Quality Data. AI systems are only as good as the data they consume. Garbage in, garbage out. Organizations that have neglected data quality—allowing duplicate records, inconsistent formats, missing fields, and outdated information—will find their AI systems generating unreliable outputs regardless of algorithmic sophistication.

Proprietary, Firm-Specific Micro-Data. As discussed in Section 3.3 of Part I of this report, generic AI trained on public macro-data provides commodity capabilities. Differentiated value requires proprietary micro-data that competitors cannot access.

Historical Depth. AI systems require time-series data to benchmark, identify trends, and learn from outcomes. Organizations that have only recently begun systematic data capture lack the historical depth that AI requires.

Integration with Decision-Making. AI insights that never reach decision-makers create no value. Organizations must invest in the reporting structures, visualization tools, and organizational processes that translate AI outputs into action.

Organizational Investment. AI deployment requires investment in data infrastructure, talent, change management, and ongoing model maintenance. Organizations that treat AI as a one-time technology purchase rather than an ongoing capability investment will see returns degrade over time.

Off-the-shelf, generic AI tools may reduce costs in narrow applications, but they do not generate competitive advantage. Organizations that expect AI to deliver transformative value without investing in these preconditions will be disappointed.

The Benefits of Valuing Intangible Assets for AI Initiatives

There is a clear need to accurately value Return on Investment in AI and consequently a need to understand the value of the intangible assets AI initiatives consume and generate. Below we detail five reasons why valuing intangible assets is essential for companies investing in AI.

How to read this section. Each of the five reasons that follow is presented using a consistent analytical structure. We begin with a short framing discussion explaining why the issue matters for AI-driven value creation, risk, or strategic positioning. This is followed by a structured summary across five dimensions:

1. Unit of Analysis: clarifies what is being evaluated (for example, a single AI initiative, a business capability, a portfolio, or the enterprise).
2. Decision Types: identifying the types of decisions the analysis is intended to inform (such as capital allocation, governance design, strategic positioning, transaction structuring, or external communication).
3. Time Horizon: indicating whether value and risk manifest in the near, medium, or long term.
4. What intangible asset valuation adds: outlines how traditional financial or operational analysis is extended by intangible asset valuation analysis.
5. Typical Metrics: potential metrics that can be used to evaluate decision outcomes.

Each reason concludes with a real-world example demonstrating how this framework changes ROI assessment, strategic choices, or value realization in practice.

Reason 1: Improving Operational Capital Allocation and ROI Accuracy

Traditional ROI measures capture only a narrow slice of value: immediate cost savings or incremental revenues that flow through the P&L.

Intangible asset valuation identifies and quantifies not only these direct benefits but also the value residing in assets created, enhanced, or degraded by an AI initiative. This includes:

1. residual value that persists beyond the project horizon (particularly data and related intangible assets that are retained, enhanced through AI, and redeployed across future initiatives); and
2. transferable value that could be licensed or sold (for example, datasets or software), and
3. option value that provides rights to expand or reuse AI capabilities elsewhere in the business.

Often these additional sources of value exceed the near-term operational benefits that initially justified the project.

It is worth noting that valuation does not allow the company to place these assets on the balance sheet. What it does, however, is provide management with critical insights into the assets that typically make up 90 percent of corporate value and are fundamental to AI value realization and risk. More and better information leads to better decisions; better decisions lead to increased value.

Summary:

1. Unit of analysis: A specific AI initiative or portfolio of comparable projects.
2. Decision types: Go/no-go, budget allocation, staging, or continuation.
3. Time horizon: Near to medium term, capturing cash returns and enduring asset formation.
4. What IA valuation adds: Assigns a financial value to assets created, improved, or degraded (e.g., data, software, regulatory approvals), allowing NPV/IRR to reflect both P&L impact and asset accretion.
5. Typical metrics: NPV adjusted for asset accretion, payback adjusted for residual IA value, risk-adjusted return.

Example: Andersen Consulting worked with a financial services firm that developed an AI claims-assessment engine projected to save $40 million annually. Our intangible asset valuation, using relief-from-royalty and with/without methods, showed the project had also created a proprietary dataset and scalable software system with at least $85 million in additional residual and transferable value. The causal pathway was: historical claims data → AI model → automated assessment capability → reduced processing costs ($40M annual savings) + proprietary dataset and software assets ($85M value). This creation of durable, monetizable assets was invisible in traditional cash-flow analysis. Recognizing this broader economic contribution reclassified the initiative from a short-term cost-reduction project to a process of capital formation in which project investment generated durable intangible assets producing future economic returns, materially changing its ROI calculus and investment priority. The organizational implications were significant: the firm established dedicated data governance to preserve and enhance the dataset and began exploring licensing opportunities for the assessment engine.

Reason 2: Accounting for Competitive Advantage and Strategic Positioning

Intangible asset valuation clarifies how AI initiatives influence competitive position—for better or worse. Many AI programs create new defensible assets such as proprietary data, algorithms, or network effects, while others may unintentionally weaken the enterprise’s competitive moat by increasing dependency on external platforms, eroding unique capabilities, or making public what once differentiated the business.

By quantifying the value and distinctiveness of key intangible assets before and after AI deployment, leaders can see whether an initiative is truly strengthening long-term advantage or quietly undermining it. This visibility is almost impossible to achieve through operational or financial metrics alone.

Summary:

1. Unit of analysis: A business capability, technology domain, or initiative where AI materially influences competitive differentiation.
2. Decision types: Where-to-play/how-to-win choices; scaling, replication, or exit of strategic assets.
3. Time horizon: Medium to long term, focused on durability and composition of competitive advantage.
4. What IA valuation adds: Forces assessment of the strength, uniqueness, and control of intangible assets, revealing where AI enhances differentiation or introduces strategic fragility.
5. Typical metrics: IA Constellation Strength (uplift or erosion), Encumbrance Ratio, Defensibility Index, IA Risk Index.

Example: Andersen Consulting worked with a global logistics firm that had invested heavily in an AI route-optimization platform built on a third-party cloud provider’s model. An intangible asset assessment demonstrated that while efficiency had improved, the initiative had substituted a previously unique in-house dataset for a generic industry one. It had also created significant dependency on external providers offering the same software and data to competitors. The causal pathway was: proprietary dataset (unique asset) → migration to third-party platform → loss of data differentiation + competitive exposure. The analysis revealed a net decline in defensible asset value. The firm had traded operational gains for long-term strategic vulnerability. Management undertook organizational changes: rebuilding proprietary data capture processes, renegotiating vendor terms, and establishing a dedicated team to reduce third-party lock-in.

Reason 3: Improved Visibility, Governance, and Risk Management

Peter Drucker famously observed that “what gets measured, gets managed.” Because most intangible assets are absent from financial statements, many organizations undertaking AI initiatives lack clear visibility of the assets those operations utilize—and struggle to manage or protect them effectively. The danger is not only that governance is incomplete, but that management may be unaware that key assets or risk exposures exist at all.

By attaching measurable economic value to these assets, organizations bring rigor and accountability to AI governance. Oversight of AI investment often concentrates on near-term cost reduction or revenue growth while overlooking medium- to long-term impacts on the integrity, ownership, and control of the underlying assets being created or relied upon. Intangible asset analysis performs a function similar to health and safety audits: cutting safety procedures to save a few dollars might make economic sense in the strictly short term, but viewed through a risk accumulation lens, those “savings” will be immaterial relative to long-term risks and costs.

Valuation enhances governance by providing more complete, higher-quality information for decision-making. This leads to stronger, more effective oversight and lower risk because decisions are made with a full understanding of the real assets at stake, not just the financial line items visible in the P&L.

Summary:

1. Unit of analysis: The portfolio of intangible assets underpinning a specific AI initiative.
2. Decision types: Risk assessment, ownership structure, compliance, or governance design.
3. Time horizon: Continuous, focused on maintaining asset integrity and regulatory readiness.
4. What IA valuation adds: Quantifies exposure, dependency, and control gaps across critical assets, enabling governance to be targeted where risk-adjusted value preservation is highest.
5. Typical metrics: Governance Risk Matrix, Encumbrance Ratio, Defensibility Index, compliance capital at risk.

Example: Andersen Consulting worked with a European bank that had deployed a generative-AI model to automate client communications. An intangible-asset review and valuation program showed that 60 percent of the model’s value was tied to a third-party large-language-model API, and that training data contained customer information subject to GDPR. The causal pathway was: customer data (regulated asset) → third-party LLM training → regulatory exposure + vendor dependency → €180M value at risk. The analysis quantified regulatory and dependency risk at €180 million value-at-risk, alerting the bank to redesign its data governance, establish internal model ownership, segregate sensitive data, and negotiate revised vendor terms. The organizational restructuring reduced both compliance exposure and ongoing oversight costs.

Reason 4: Enhanced Transaction Value and Deal Structuring

Because intangible assets are largely absent from financial statements, the economic value embedded in AI capabilities can be invisible to potential buyers or investors. In transactions involving AI-rich businesses or technologies, the seller must identify and articulate the value of underlying assets. Without that evidence, the counterparty will either not see or not acknowledge their contribution, and the price will often default to conventional earnings multiples that ignore much of what is truly being exchanged.

Proper valuation shows why these assets may be worth substantially more than short-term cash flow models suggest, particularly where AI capability is strategically complementary to the acquirer—the key that fits their specific lock. This goes directly to the issue of Indirect AI Value Creation addressed in section 1.2.

By quantifying these assets, valuation reframes negotiations around demonstrable, valuable assets rather than shorter-term economic returns that may not accurately reflect the true value of the enterprise. This leads to more accurate pricing and better deal terms.

Summary:

1. Unit of analysis: The enterprise, AI business unit, or technology being transacted.
2. Decision types: Acquisition pricing, spin-out valuation, licensing, or earn-out design.
3. Time horizon: Event-driven, realized at transaction but with enduring implications for value capture.
4. What IA valuation adds: Quantifies asset composition and contribution, substantiates value claims, and supports or amends multiples in negotiation.
5. Typical metrics: Allocated IA value by class, proportion of enterprise value attributable to AI-driven assets, license-royalty equivalence value, synergy-adjusted asset value.

Example: Andersen Consulting worked with a mid-sized enterprise software vendor being acquired by a global systems integrator. The seller’s financials showed €12 million EBITDA, supporting a €100 million valuation on standard multiples. Intangible-asset valuation demonstrated that a proprietary AI orchestration layer, built upon internally generated data and algorithms, had unique strategic value to the acquirer, who would embed it across a global client base. The analysis contributed €70 million of strategic intangible value, leading to a higher purchase price and more favorable earn-out terms. The causal pathway was: proprietary AI layer + unique data → strategic fit with acquirer’s global distribution → enhanced value realization → €70M additional strategic value.

Reason 5: Making the Valuable Visible: Investor and Stakeholder Communication

Value that remains unseen, unmeasured, or unarticulated contributes little to confidence or decision-making. Investors, boards, and regulators see the costs (large R&D investments) and potentially the direct anticipated benefits (future income streams) but not the assets being created.

Intangible asset valuation makes this “invisible” AI value visible. It provides management with tools to explain, in financial terms, how AI investment translates into enduring intangible assets such as new data, codified know how or regulatory approvals. This allows leadership to communicate not only that investment in AI is occurring, but why it is strategically and economically justified.

By telling this story clearly, management shifts the perception of AI from expense to capital formation, from speculative to evidence-based. This strengthens confidence among investors and internal stakeholders alike, aligning perceptions of performance with the underlying economics of value creation.

Summary:

1. Unit of analysis: Aggregated AI investment program or business-unit portfolio.
2. Decision types: External disclosure, valuation modeling, or investor communication.
3. Time horizon: Medium term, aligned with capital-market reporting and valuation cycles.
4. What IA valuation adds: Translates AI investment into identifiable, durable assets, bridging the gap between R&D expense and balance-sheet growth and enabling credible, data-backed communication.
5. Typical metrics: Total IA value created, ROIAI (Andersen Proprietary Metric), enterprise-value uplift from intangible-asset growth, proportion of AI investment capitalized as intangible formation.

Example: Andersen Consulting worked with a listed insurer who had spent approximately €60 million on AI R&D. Intangible-asset valuation identified €170 million in newly created software, data, and regulatory assets. Communicating this to investors reframed AI not as a cost center but as a driver of capital formation and sustainable enterprise value, improving investor confidence and leading to a material improvement in the stock price. The causal pathway was: R&D expenditure → identifiable intangible asset creation → investor communication → improved market confidence → material stock price improvement. The organizational implication: the firm established quarterly intangible asset reporting to its board and initiated proactive analyst engagement on AI value creation.

Towards a More Effective Model of ROI Calculation

The preceding section explained why valuing the intangible assets underpinning AI initiatives matters. This section provides a practical illustration of how that valuation can be achieved.

The Andersen Consulting Strategic Intangible Asset Value Accretion Model (SIAVAM) is one of several proprietary frameworks we have developed for quantifying the economic impact of AI initiatives. It enables executives to view AI investment through both a P&L and an asset lens, converting an invisible or qualitative story into a structured, evidence-backed financial analysis. It is designed to sit alongside and complement traditional financial analysis tools.

SIAVAM explicitly recognizes that intangible assets are critical inputs into AI and that AI initiatives generate valuable intangible outputs. The model:

1. Identifies valuable intangible assets relevant to the organization’s operations and strategy;
2. Estimates the value of these assets before the AI initiative begins;
3. Identifies new intangible assets created as a result of the AI initiative;
4. Measures and quantifies the value of these newly created intangible assets, specifically accounting for their long-term impact;
5. Assesses pre-existing intangible assets for any value change (up or down) caused by the AI initiative;
6. Calculates the potential increase or decrease in the value of the entity’s intangible assets as a function of investment in AI.

In doing so SIAVAM explicitly captures the value of residual intangible asset creation through the AI flywheel, which is ignored by conventional payback-style ROI methods.

For example, an internally generated dataset sitting off balance sheet can be identified and valued. The outputs from the AI process utilizing that dataset—such as a new software system—can also be valued. The resulting increase in the dataset’s value can also be quantified. All of these value bearing elements are ignored in traditional P&L-style ROI calculations, which overlook the critical role of asset utilization, creation, and modification.

Illustrative Case: Multinational Insurance Company

Andersen Consulting was engaged by a large multinational insurance company that reported $98.1 million in R&D expenses related to a key AI initiative. When analyzed using our SIAVAM model, these “expenses” were found to have generated approximately $298.6 million in asset creation:

Intangible Asset	Strategic Role	Expense Incurred	SIAVAM Value Estimate
Software Stack	Orchestrates automation, scales delivery, reduces cost-to-serve	($38.2M)	$112.0M
Customer Data	Powers improved targeting, retention, personalization, driving revenue	($20.9M)	$70.8M
Regulatory Approvals	Enables barriers to entry, pricing power, ensures compliance	($24.3M)	$65.6M
Processes & Systems (Industry Expertise)	Enables scalable, repeatable delivery and operational efficiency	($9.4M)	$38.6M
Content	Preserves institutional knowledge and reduces performance variability	($5.3M)	$11.6M
Total		($98.1M)	$298.6M

Table 1: Step by step valuation of intangible assets through the SIAVAM valuation model.

This insight enabled management to:

1. Reframe discussions with internal stakeholders in terms of Return on Invested Capital rather than expenditure. Utilizing our Return on Intangible Asset Investment (ROIAI) framework, management demonstrated a strong return on capital rather than presenting an essentially meaningless expense metric.
2. Engage with analysts and explain how AI investment would drive enhanced competitive advantage over the medium term. Analysts were persuaded by this analysis and accompanying narrative. Over the subsequent twelve months, the stock climbed approximately 60 percent, materially outpacing the broader index and generating alpha.

The causal pathway – project-level investment generating data and other intangible assets, those assets being enhanced through AI, and the resulting future cash flows – was now explicit and traceable. The organizational implications were substantial: the firm restructured its AI governance, established formal intangible asset tracking, and reallocated capital toward initiatives with stronger intangible asset formation profiles.

While SIAVAM-based analysis cannot be used to revise statutory accounts, it can and should be applied to management decision-making and stakeholder communications. It provides a powerful view of the economics of AI initiatives that has previously been unavailable. It enables decision-makers to see “the other side of the coin” and integrate the asset dimension of AI into their ROI calculus. Crucially, the first step in the valuation process is the systematic identification of the relevant intangible assets. This means that even independent of the final valuation outputs, SIAVAM materially strengthens governance and strategic control by making these assets visible—enabling clearer ownership, improved oversight of risk, and more disciplined capital-allocation decisions. This directly supports the objectives outlined in Reasons (2) and (3) in Section 2. In this way, SIAVAM operates not only as a valuation methodology but also as a governance framework: it quantifies the economic contribution of AI while simultaneously revealing, structuring, and reinforcing the management of the intangible assets on which that value depends.

The Relationship Between Intangible Asset Valuation and Cash Flow

The devil’s advocate or realpolitik investor may argue: “Yes, it’s great to value intangible assets and understand their governance, risk, and competitive advantage implications and even to use them to improve stakeholder communications —but none of this changes cash flow, and cash flow is ultimately what investors care about.”

This is not completely correct.

It is true that many valuation models rely on cash flow. There is a good reason for that: cash flow is an excellent and reliable measure. It also goes to the core of management, stakeholder, and investor concerns: “If I put this much cash in, how much cash will I get back?” This question lies at the heart of most investment decisions.

It is also true that understanding and valuing intangible assets to enable better governance, enhanced risk management, increased competitive advantage, and improved stakeholder communication does not, in itself, immediately alter cash flow. In this sense, valuing intangible assets is akin to taking a photograph of a business: the act of taking the photo does not on its own change the scene it captured.

But while the act of taking the photo does not change the scene, it can change behavior of the participants or the photographer if they see things they might otherwise have missed. Given that (1) intangible assets are essentially outside financial statements and (2) are highly consequential to enterprise performance and growth, there is a very real probability that closer inspection of the photograph (the valuation) will change participant behavior—and hence enterprise value.

This goes to the heart of the benefits outlined in Section 2. Identifying and valuing intangible assets that drive AI initiatives provides management, stakeholders, and investors critical intelligence to adjust behaviors or decisions. It provides more and better information. And ceteris paribus, more and better information means better decisions and increased value.

If, as CFO, I know that an AI initiative will increase the value of my existing dataset, that is relevant to my decision to proceed. If, as a director, I know that the AI initiative may open the company to a major patent lawsuit, that is relevant to my risk assessment. If, as an investor, I realize that an acquisition target’s AI infrastructure can be massively scaled under new ownership, that too is relevant to pricing—and hence to value.

In short, even if none of these scenarios change short-term cash flows, this additional intelligence is highly valuable and may change assessments of medium- to long-term cash flow, thereby indirectly altering enterprise value. Or, in the case of identifying new intangible assets or revaluing existing ones, directly impact enterprise value.

Decisions taken as a consequence of the additional information provided by valuation may lead directly to changes in:

Cash flow — An AI initiative that drives cost reductions that would not have been advanced is funded because it also creates valuable data.

The company’s cash position — The company sells a business unit for more than discounted cash-flow value because it can point to the existence of powerful AI algorithms valuable to the acquirer.

The market’s assessment of company value — The business communicates to investors the value of a new AI content-creation engine, and the market responds favorably.

In summary: intangible assets are central to AI value creation. Understanding their value provides managers, stakeholders and investors with better information. Better information enables better decisions. Better decisions create more value.

Conclusion and Implications

The Unified Argument

Artificial Intelligence and intangible assets are inseparable. AI consumes, transforms, and produces intangible assets in a continuous cycle. Understanding this relationship is essential for any organization seeking to measure the true economic return on AI investment.

Parts I and II of this paper have advanced three interconnected arguments:

First, intangible assets constitute the dominant source of modern enterprise value—representing approximately 90 percent of S&P 500 value—yet they remain largely invisible in financial statements. This invisibility creates a structural blind spot: the assets that generate cash flows and competitive advantage are off balance sheet, unrecorded, and unvalued. Modern accounting reveals what is visible but not what is valuable.

Second, this blind spot has concrete, material consequences. Without visibility into intangible assets, organizations cannot allocate capital effectively, manage risk appropriately, or communicate value credibly. They invest heavily in the foundations of their competitiveness without recording them as assets or understanding their economic return. Correspondingly, while AI initiatives are often evaluated as discrete projects with short payback horizons, their true economic impact lies in the residual intangible assets they create—particularly data—which can be retained, enhanced through AI, and reused to generate compounding future cash flows.

Third, AI both depends on and transforms intangible assets. AI requires proprietary micro-data, accumulated over time, to generate differentiated insights. It demands organizational changes to data capture, storage, reporting, and decision-making. And it produces new intangible assets—software, data, know-how, regulatory approvals—that compound competitive advantage. The relationship is not static but dynamic: a flywheel in which each rotation compounds the firm’s advantages or accelerates its decline.

Implications for Organizations

Organizations that recognize these dynamics and act accordingly will outperform those that do not.

Specifically:

Invest in Data Infrastructure. The flywheel depends on proprietary micro-data captured systematically over time. Organizations must redesign data capture processes to generate clean, structured, timestamped data. They must invest in storage architecture capable of handling the volume and variety that AI requires. And they must treat data not as an operational byproduct but as a strategic asset whose value compounds through time.

Prevent Data Degradation. Every day of unsystematic data capture is a day of permanent value destruction. Organizations must address the mechanisms through which data degrades: non-capture, decay, and departure of tacit knowledge. AI systems can serve as a hedge against degradation—but only if organizations invest accordingly.

Adapt Organizational Structures. AI is not a tool bolted onto existing operations. It is a transformation engine that restructures how firms operate. Organizations must adapt reporting structures to surface AI insights to decision-makers. They must redesign roles to accommodate human-AI collaboration. And they must develop governance mechanisms appropriate to the intangible assets AI creates and consumes.

Identify and Value Intangible Assets. Organizations cannot manage what they cannot see. The first step toward effective AI governance is identifying the intangible assets the organization possesses, the assets it creates through AI initiatives, and the assets upon which AI effectiveness depends. Valuation provides the visibility necessary for informed capital allocation, risk management, and stakeholder communication.

Consequences for Organizations That Fail to Act

Organizations that fail to recognize and act on these dynamics face predictable consequences:

Competitive Disadvantage. Competitors who understand their intangible asset base and manage the AI flywheel will build increasingly insurmountable leads. Each rotation of the flywheel compounds their advantages.

Misallocated Capital. Without visibility into the assets AI consumes and creates, organizations will systematically misallocate capital—underinvesting in initiatives that build strategic capability and overinvesting in those that provide only near-term operational gains at the cost of long-term intangible asset erosion.

Degraded AI Effectiveness. Organizations that have not systematically captured and retained proprietary micro-data lack the fuel AI requires. Their AI systems will generate commodity outputs indistinguishable from competitors’—or worse, will underperform due to data quality and completeness deficiencies.

Undervaluation. In transactions, even organizations with strong intangible asset bases that cannot identify and articulate the value of those assets will be systematically undervalued. Value that cannot be seen cannot be priced.

Strategic Vulnerability. Organizations that fail to proactively assess the encumbrances against their AI initiatives—patent exposure, regulatory risk, third-party dependency—will discover these vulnerabilities too late – when cash flows fail to materialize or when adverse events occur.

The Path Forward

The path forward is clear: organizations must bring intangible assets into view. They must understand what assets they possess, how those assets fuel AI, what assets AI creates, and how the flywheel can be managed to compound competitive advantage.

Through Andersen Consulting’s frameworks, organizations can now quantify these hidden assets. Our models provide a structured, evidence-based way to capture and value both sides of the AI investment equation: the cash flow effects visible in the P&L and the asset effects traditionally excluded from it. By revealing and valuing the intangible asset base, organizations strengthen governance, improve strategic decision-making, and provide a credible foundation for stakeholder communication.

Those that take this step will move beyond guesswork to demonstrable, compounding enterprise value creation. Those that do not will find themselves progressively disadvantaged in an economy where intangible assets and AI are inseparable—and where managing their intersection is the central challenge of strategic leadership.

---

©2026 Andersen Institute for Finance & Economics. All Rights Reserved. This material is confidential intellectual property of the Andersen Institute for Finance & Economics. The views expressed in this note are those of the authors and do not represent an official position of The Andersen Institute for Finance and Economics or affiliated organizations.  By viewing this Andersen Business Insight, you agree that you will not directly or indirectly copy, modify, record, publish, or redistribute this material and the information therein, in whole or in part. No warranty or representation, express or implied, is made by the Andersen Institute or any of its affiliates, nor does Andersen accept any liability with respect to the information and data set forth herein. Distribution hereof does not constitute legal, tax, accounting, investment or other professional advice. The information provided herein is not intended to provide a sufficient basis on which to make an investment decision. Recipients should consult their own advisors, including tax advisors, before making any investment.

[1] Paul Adams and Jason Strimpel are managing directors at Andersen Consulting.  This paper has benefited greatly from very helpful comments, insights and suggestions from Myron Scholes.