CEO and cofounder at Open Trellis.
XAI, or eXplainable Artificial Intelligence, presents business leaders with new opportunities. XAI differs from so-called black box AI in terms of transparency. Claude, a publicly accessible AI, submits, “Explainable AI actively aims to make AI transparent, while black box AI is by definition opaque and unexplainable.” XAI promises more solutions of strategic importance across profit centers.
AI is not sentient, nor is it inventive. Its creativity is combinatorial, as it can only return combinations of pre-input data. Its predecessors, like Computer Generated Imagery (CGI), were first created in 1958, and T9 predictive text was invented in 1995. The recent advancements in large language models (LLMs) have brought about longer-form results, as well as conversational style inputs and results.
Consumer-facing AI is also returning image and video outputs. XAI expands potential business applications, as it allows evaluation and audit of inputs, process structure, output rationale and increases in system accuracy.
The Opportunities Emerge
XAI’s emergent business opportunities include:
• Cost optimization through integration with quantitative analytics and resultant applications across functions—accounting, finance, operations, risk management, marketing and strategic planning.
• Productivity improvements through voice-based input enhancements, especially in the efficiency of content creation.
• Image recognition, evaluation and human decision support.
• New product research and evaluation through rapid customer engagement, concept designs, prototyping and commercialization with trial users.
• Integration with adjacent technologies, such as additive manufacturing, quantum security and blockchain.
However, risks abound. Associated programs that apply XAI within core and support functions can face stiff resistance. Efficiency gains often mean layoffs. Second, evolutionary programs should be favored over revolutionary business cases. Third, the creation of a technology does not ensure sufficient adoption and diffusion (registration required).
While there is a cost of market share loss when maintaining the status quo, projects are evaluated in light of potential cost overruns. Conventional project finance still applies. XAI investment costs can be mitigated through the use of consumer-based tools, API development, open innovation and consideration of collaborative solutions.
Example: Integrated Technology In Debt Financing
Dr. Swati Sachan and I have published research on applications of blockchain and XAI to assist human underwriters in evaluating business loan applications. We have extended our research to include the DeFi (Decentralized Finance) Trustboost framework, which includes combinations of blockchain technology and XAI within a multi-layer architecture.
The framework aggregates the knowledge on lending rules for underserved communities from multiple financial experts to facilitate financial inclusion. The aggregated information is utilized as input data in an explainable deep neural network model. The architecture of this model includes a depth of layers and nodes; it is feed-forward (i.e., flows from input to output). Critically, the model’s decisions are interpretable, empowering human underwriters to maintain an active role in decision making to enable continuous advancements in algorithmic decisions driven by the correction of output errors.
The architecture is designed to maintain confidentiality, ensure resilience to adversarial attacks, comply with data protection laws, enable regulatory audits and optimize on-chain versus off-chain computational costs.
How To Identify XAI Opportunities
Beyond cost reduction, strategic growth opportunities are found at the intersection of business advantage and customers’ demand. Opportunity identification extends beyond technological advancements and covers geopolitical, demographic, social, regulatory and environmental trends. Current economic considerations are mixed but help illuminate context:
• U.S. Gross Domestic Product is growing at 2.4% annualized.
• Inflation is cooling, with indices showing a slowdown in price increases.
• Unemployment is low, and wage levels remain strong.
• The Fed Funds Rate target reaches 5.25 to 5.5%, the highest level in 22 years.
• Bank vulnerability persists, especially for smaller regional lenders, and commercial real estate (paywall) is under pressure, particularly in soft demand for office spaces.
• Increases in corporate loan defaults and in consumer bankruptcies.
• Almost all of the past 12 month’s gains in the S&P500 have been driven (paywall) by the “Big 5 Tech Companies.”
Industry-level evaluations of XAI implications start with traditional analyses of risk-adjusted returns, structural competitive intensity and entry barriers. Inorganic growth can be achieved through target acquisitions in attractive industries. These firms are developing complementary XAI solutions or providing pathways to new beachhead markets. XAI-related cost efficiencies will create downward pricing pressure within those industries that are most price competitive.
Ask yourself: Is the target firm available at an appropriate valuation? Are there strategic synergies? To what extent can we hold market share and sustain margin? Can XAI expand total projected cost savings in a post-acquisition scenario?
Product-level assessments for organic growth can yield further opportunities for both new ventures and established firms. Opportunities span the saturation of current segments with existing offerings, selling to new segments and commercializing new products.
• What new XAI products, or XAI-augmented services, are being introduced within core markets? At what rate are customers shifting purchases?
• How might our current business development strategy be enhanced by utilizing XAI for lead generation, refining our awareness and engagement campaigns, improving first-glance UX, converting higher percentages or enhancing post-sale services?
• Which disruptive products/services have changed the purchase patterns of our current customer segments? What new benefits did they start receiving? How might that change spur a new purchase pattern in our markets?
• How might XAI assist in developing marketing content that attracts new segments to our existing portfolio? How does this dynamic change based on geography, demography and psychographics?
• Might new customer segments be won over through incremental XAI enhancements within our current portfolio?
• How might we utilize XAI-enhanced R&D to increase our speed to commercialization, the quality of our current products, or reimagine concept development?
XAI has the potential to spur cost reduction, enhancements to product development, other top line growth opportunities and transformational business models. Business leaders are assessing these opportunities through the evaluation of technologies, the pace of their advancement and adoption, industry dynamics, project risks and implications across their current and prospective product markets.
Firms developing XAI are rightly focusing on strong use cases for business implementation and value creation. It’s still worth evaluating the investment opportunities through a finance lens. Think Net Present Value and Internal Rate of Return.