In finance, asset managers oversee a range of valuable entities, from liquid assets like cash to tangible investments such as property. AI solutions, whether they enhance operational efficiency or amplify sales, embody intrinsic monetary value for businesses. It is important for corporate leaders to view their AI tools not merely as technological utilities but as integral assets within their area of responsibility.
In this article we draw analogies between managing a securities portfolio and overseeing an AI suite. Much like a financial portfolio, a company’s collection of AI applications is built progressively. Each component has its risk profile and necessitates oversight to ensure continued value delivery. Notably, the distinctive risk characteristics of AI can provide diversification, acting as a strategic hedge hedge within the broader asset mix.
A genuine business need starts the discussion of making an AI addition to the portfolio. When demand is recognized, the potential use case is thoroughly articulated, evaluated for feasibility, and assessed for its contribution to the business’s bottom line. Should a formal decision-making body and process be in place, the proposed AI initiative would go through these gates for validation, as detailed in our prior issue on use cases and their value generation.
AI solutions, which we think of as the technological response to business challenges, hold merit only when they align with and advance the company’s strategic goals. Echoing the sentiments of a seasoned compliance officer: the essence of business lies in weighing opportunities against the associated risks. This is particularly pertinent for AI, where its potential to unlock business prospects is often recognized but whereas leaders are still learning the intricacies of counterbalancing inherent risks. As stewards of their company’s AI asset portfolio, leaders must evaluate these factors to pursue ventures that promise the most significant return.
Benefiting from the financial upside of AI investment
AI technologies only qualify as assets when they serve your company’s strategic goals. A common pitfall is the hasty development of AI tools in search of problems, rather than the reverse and such an approach often leads to wasted resources. It is crucial to anchor AI initiatives in concrete business requirements, ensuring that the development of AI solutions is purpose-driven and aligns with identified needs.
A mere six months following the public revelation of the Chat-GPT innovation, experts predicted that generative AI could potentially boost global GDP by an estimated 7% within the forthcoming decade [1]. Given that business growth generally parallels economic expansion, participation in AI becomes essential to benefit from this prospect. Similar to the dynamics of the financial markets, a lack of investment in a new asset class like AI equates to forgoing its potential gains. Inclusion of and strategic allocation in AI are key to securing a share in this growth trajectory, much like a smartly diversified investment portfolio.
To maximize benefits of AI, businesses are advised to adopt a calculated approach to integration. This involves not only identifying areas where AI can drive efficiency and innovation but also investing in the necessary infrastructure and talent to deploy these solutions effectively, positioning your organization to gain a competitive edge. Ongoing investment in AI capabilities can compound over time, resulting in a significant advantage as these technologies evolve and their applications expand. As with any strategic investment, the key to success lies in recognizing the potential and to subsequently commit to nurturing gradual growth within the organization.
Navigating the risks in AI asset management
Risk management is as crucial as seizing opportunities. The risks of AI exposure — and notably, the risks of non-exposure — must be acknowledged and addressed. Neglecting AI investment can lead to loss of market share, as competitors harness AI to resolve issues previously deemed too resource-intensive. A proactive approach to AI can bring cost-efficient solutions, giving businesses a competitive advantage in areas once considered intractable.
Although AI-related risks could be categorized in many different ways, let us here distiguish between two main types. Firstly, AI-idiosyncratic risks pertain to the specific challenges of implementing and operating AI systems, such as algorithmic bias, data privacy concerns, and the unpredictability of AI behaviors. Secondly, there are the typical risks associated with pursuing new business opportunities, such as misalignment with corporate strategy, market reception, and uncertainties related to the return on investment.
Conversely, AI solutions can also serve as a hedge against a spectrum of operational risks. For instance, AI can mitigate the impact of a limited skilled labor pool by automating some tasks, assist in employee and customer retention through predictive analytics. Moreover, AI can equip businesses with advanced dynamic pricing strategies, offering a defense against price competition in the market.
A critical and unfortunately typical oversight on the AI journey is the presumption that advances like Chat-GPT eliminate the need for robust and business-friendly data modeling or even high-quality data. The quality of AI outputs is still heavily dependent on the quality of the inputs and therefore, maintaining a strong foundation in data governance is a prerequisite for AI to be a reliable and effective tool in the corporate toolkit.
A concrete example – AI for ESG performance and reporting
In today’s business world, an AI portfolio is incomplete without solutions addressing environmental, social, and governance (ESG) concerns. With global attention shifting towards sustainability, companies are now required to provide detailed reports on environmental impact and ensure transparency within their supply chains [2]. At its core, this challenge is rooted in data management, where AI can play an important role.
In markets like Finland, financial incentives underscore the importance of ESG compliance; for example, a leading bank offers favorable business loan terms to companies meeting sustainability benchmarks [3][4]. Artificial intelligence, especially generative models, is adept at synthesizing vast amounts of unstructured data into coherent datasets essential for accurate sustainability reporting.
Concluding thoughts: Prudent AI asset management in business
The AI solution suite, much like a diversified securities portfolio, is a strategic asset overseen by business leaders. These solutions have both the potential for significant returns as well as exposure to risk. Early projections suggest that generative AI could elevate global GDP by 7% within a decade, with potential productivity growth possibly of 1.5% in that same period [1]. In an era where efficiency is on most corporate leaders’ agenda, AI investment has quickly been established as an important option for companies to explore for both those growth and productivity increase opportunities.
Realizing value from an AI portfolio relies on the relevance of the AI solutions to real business challenges. While some organizations may encounter disappointing return on investment from their AI initiatives, often this comes from a disconnect between data science and business objectives. It’s crucial to recognize that tools like Chat-GPT, while innovative, do not automatically secure a competitive edge. That advantage is only achieved when AI is tailored to leverage a company’s unique data assets effectively.
Regular portfolio assessments and proactive rebalancing are essential practices, as it ensuring that AI systems have access to quality data and that they align with evolving business needs. Like financial portfolio management, strategic choices in AI investments must be made with careful consideration. As new AI opportunities present themselves, it may be necessary to decommission older solutions to allocate resources to more promising ventures. The focus should always be on selecting the AI assets that promise the most substantial impact on the company’s performance.
Terminology
AI-idiosyncratic risk – a type of risk that is particular to AI solutions as opposed to commonplace business risk. (Here we use terminology from portfolio theory, where risks due to the unique circumstances of a specific security are called “idiosyncratic risks”.)
Algorithmic bias – refers to systematic and repeatable errors in an AI solution (or computer system)s that create unfair outcomes or make decisions that systematically disadvantage certain groups of people.
Author: Hanna Grönqvist
References
- Goldman Sachs “Generative AI could raise global GDP by 7%” (5th April 2023)
https://www.goldmansachs.com/intelligence/pages/generative-ai-could-raise-global-gdp-by-7-percent.html
- European Commission – Corporate sustainability reporting
https://finance.ec.europa.eu/capital-markets-union-and-financial-markets/company-reporting-and-auditing/company-reporting/corporate-sustainability-reporting_en
- European Investment Bank’s InvestEU program https://ec.europa.eu/commission/presscorner/detail/en/ip_21_1046
- Green business loans by major Finnish bank
https://www.nordea.fi/en/business/our-services/financing/verde-programme.html