AI in Finance and Banking, April 15, 2024

This semi-monthly column highlights news, government documents, NGO/IGO papers, industry white papers, academic papers, conferences and speeches on the subject of AI’s fast paced impact on the banking and finance sectors. The chronological links provided are to the primary sources, and as available, indicate links to alternate free [unpaywalled] versions.


City AM, April 15, 2024. City leaders turn to AI bots to fight financial crime. Half of leaders in the City of London use AI to access financial advice, according to new research. KPMG UK’s figures also show that 36 per cent of senior leaders in the financial services sector use the tool, with chief executives the most frequent users at 30 per cent. According to the firm, CEOs use it up to three times a day mostly for brainstorming (50 per cent) and help with speech writing and presentations (46 per cent). KPMG UK added that despite more than 60 per cent of leaders using generative AI at least once a week in their day jobs, almost a third (31 per cent) aren’t confident that the business overall is harnessing the potential of the technology.  According to the data, financial planning is where most leaders (44 per cent) say generative AI is being used in their businesses, followed by customer data analytics (38 per cent) and fraud detection (35 per cent).  KPMG Uk said that over the next three years, most leaders call out financial planning (40 per cent), marketing (37 per cent) and fraud detection (36 per cent) as where the tech will be used. Using generative AI to help fight financial crime aligns with concerns related to its use for criminal intent, as highlighted in another recent KPMG study, which showed that this is the biggest concern amongst UK adults.

Forbes, April 14, 2024: AI For Venture Capitalists: Trends To Watch And Key Considerations. There’s little doubt artificial intelligence is changing how people run their businesses, develop technology and make investments. While some jobs might be impacted negatively, I believe that AI will create opportunities and have a net positive impact on the economy. AI can help invent new technologies, make the world more productive and, ultimately, positively impact the world. According to a PwC report, AI could potentially contribute $15.7 trillion to the global economy by 2030. By then, the report states, local economies worldwide are expected to grow up to 26% due to AI. This growth is expected to be driven primarily by product enhancements driving increased customer demand, accounting for 45% of the economic gains. This explosion in AI presents a major opportunity for venture capital firms. Based on my experience investing in this space, here are some of the trends VCs should watch, along with key considerations that are important when vetting an AI startup investment.

CNBC, April 14, 2024. It’s not just Jamie Dimon and Wall Street. Local bank branches have big AI ambitions. The pandemic accelerated changes at big banks, where Chase and Wells Fargo already have branches that look more like lounges than banks. But it’s not just Wall Street-sized banks where AI is disrupting the way things works. Small, independent branches are also following, and experts and executives say they’ll use their small size and agility to their advantage. The local bank branch, with its traditional teller windows and long lines, will transform into an AI-infused, customer-centric financial services center, aiming to beat the big banks on the service that AI will allow them to provide customers. “As a small bank, your only value proposition is service. Nothing is proprietary anymore,” said Christopher Naghibi, executive vice president and CEO of Irvine, California-based First Foundation Bank, which has 43 branches in five states. With just over $10 billion in assets, Naghibi helped shepherd First Foundation from a single branch in 2007 to its size today. Naghibi envisions community bank branches with fewer employees and more AI. The employees would be freed to help customers reach their financial goals and not be stuck answering basic questions about recent transactions and account information. “The teller line, as we see it today, will eventually die,” he said. Naghibi isn’t alone among bank CEOs contemplating the AI future for financial workers and customer interactions. Jamie Dimon, the veteran chairman and CEO of JPMorgan Chase, has written about artificial intelligence in his annual shareholder letters dating back to 2017. But his latest letter, released on Monday, was notable not only for his AI predictions — he wrote it could be as transformational as the printing press, the steam engine, electricity, computing and the internet — but also how he thinks the technology could impact the jobs of the bank’s more than 310,000 employees.

TechTimes, April 11, 2024: AI Could Soon Replace Entry-Level Wall Street Analysts. AI analysts over college graduates. Artificial intelligence is reportedly being leveraged by Wall Street firms to potentially help and/or replace employees with certain workloads, with entry-level analysts immediately at risk, as reported by the New York Times. Wall Street and investment banks, who have been accustomed to cultural shifts for a while, are starting to embrace generative artificial intelligence as a burgeoning technology that has the potential to replace entire workforces rather than just augment existing ones.

Fast Company, April 11, 2024. CFOs are cautiously experimenting with AI. Here’s what that means for finance teams The technology has potential to speed up financial planning and analysis—and help automate tedious tasks like processing invoices. Slowly but steadily, artificial intelligence is coming to corporate finance departments. CFOs and their teams are in most cases still exploring the capabilities of generative AI, often using it for help with tasks drafting documents and transcribing meetings so a human employee doesn’t have to take notes. “Nobody does that anymore if you’ve got Copilot,” says Kirstie Tiernan, a practice leader at IT firm BDO Digital, referring to Microsoft’s line of AI assistants. Tiernan says she’s seen businesses using AI tools to generate material for risk management practices, like documents and videos for a cybersecurity incident simulation. And software vendors catering to finance teams, like financial reporting platform Workiva, are increasingly rolling out AI features of their own.


CRS Report – Artificial Intelligence and Machine Learning in Financial Services, April 3, 2024: The financial industry’s adoption of artificial intelligence (AI) and machine learning (ML) is evolving as financial firms employ ever greater levels of technology and automation to deliver services. Expanding on earlier models of quantitative analysis, AI/ML has often been adopted in finance to solve discrete challenges, such as maximizing profit and minimizing risk. Yet the industry’s adoption of the newer technology also occurs against perceptions that are steeped in tradition and historical financial regulation, and regulators want to ensure that the technology does not sidestep regulations frequently described as technology neutral. Technological advances in computer hardware, capacity, and data storage—which permit the collection and analysis of data—helped fuel the development and use of AI/ML technologies in finance. Unlike older algorithms that automated human-coded rules, new AI models can “learn” by themselves and make inferences and recommendations not identified by modelers in advance. This shift in technology has also enabled the use of new types of data including alternative data (i.e., data that the consumer credit bureaus do not traditionally use), unstructured data (images or social media posts, etc.), and unlabeled information data—which, when combined, extend the technologies’ uses to new financial services or products. Different parts of the financial services industry have adopted AI/ML technology to varying degrees and for various purposes. Some uses of AI/ML include powering chatbots in customer service functions, identifying investment opportunities and/or executing trades, augmenting lending models or (more sparingly) making lending decisions, and identifying and preventing fraud. The extent to which a sector or firm adopts various technologies reflects a variety of factors, including a firm’s ability to fund internal development and regulatory requirements. The increased use of AI/ML to deliver financial services has attracted attention and led to numerous policy issues and subsequent policy actions. Such policy actions culminated in (1) the establishment of a task force on AI in the 116th Congress and the more recent working group in the House Committee on Financial Services in the 118th and (2) 2019 and 2023 executive orders. The evolving legislative and regulatory framework regarding AI/ML use in finance is likely, at least in part, to influence the development of AI/ML financial services applications. Various financial regulators have indicated that regulated entities are subject to the full range of laws and regulations regardless of the technology used. Additionally, some regulators have identified regulations and issued guidance of particular relevance to financial firms employing AI/ML technologies. Financial industry policymakers face competing pressures. Financial service providers and technology companies are likely to continue adopting and promoting AI/ML to save time and money and promote accessibility, accuracy, and regulatory compliance. However, challenges and risks in the form of bias, potential for systemic risk and manipulation, affordability, and consequences for employment remain. Determining whether the existing regulatory structure is sufficient—or whether one that is more closely tailored to the technological capacities of the evolving technology is necessary—has emerged as a key consideration. Should Congress consider the legislative framework governing AI/ML in finance, industry and consumers alike will expect that it weighs the benefits of innovation with existing and potential future challenges and risks.


SSRN – Fedyk, Anastassia and Kakhbod, Ali and Li, Peiyao and Malmendier, Ulrike, ChatGPT and Perception Biases in Investments: An Experimental Study (April 8, 2024). Available at SSRN. Applications of artificial intelligence (AI) in finance have been met with concerns about algorithmic bias, following issues observed in domains such as medical treatment and lending. We ask whether AI models accurately capture investment preferences across demographics. We elicit investment preferences from over 1,200 survey participants and compare the data directly to investment ratings generated by OpenAI’s ChatGPT (GPT4). We find that ChatGPT predicts investment preferences with high accuracy across demographics. Specifically, ChatGPT correctly predicts that women rate stocks lower than men, older individuals prefer holding cash, and higher incomes are associated with higher ratings for stocks and bonds. Moreover, free-form responses from ChatGPT focus on the same aspects as human free-form responses. Most common themes in both responses are “risk” and “return,” and “knowledge” and “experience” play an important role for stock market participation. One difference is that ChatGPT responses are almost always transitive, whereas human responses are more prone to violating transitivity, especially when expressing indifference. Overall, the use of AI in finance offers a promising direction for augmenting human surveys in preference elicitation, with important applications for areas such as robo-advsing.

NBER – Economics of Artificial Intelligence: Political Economy ChallengesToronto, Canada – September 19-20, 2024. The eighth annual NBER Economics of Artificial Intelligence (AI) Conference will be held on September 19-20, 2024 in Toronto, Canada. The conference is made possible by the generous support of the Alfred P. Sloan Foundation. The conference will focus on the political economy of artificial intelligence. It will be co-organized by NBER researchers Ajay Agrawal, Joshua Gans, and Avi Goldfarb of the University of Toronto, and Catherine Tucker of MIT.  Papers from the seventh conference are available.

Topics related to the political economy of AI include, but are not limited to:
* AI and political power
* AI and misinformation
* AI’s impact on public finance
* AI and the economics of national security
* AI and comparative advantage * AI and the economics of surveillance

Original research papers, as well as overviews that offer a research agenda, may be presented. The NBER especially welcomes submissions by scholars who are early in their careers, who are not NBER affiliates, and who are members of under-represented groups. To be considered for inclusion on the program, upload completed papers by 11:59pm (EDT) on June 1, 2024. Please pass this call for papers on to colleagues who might be interested. Please do not submit papers that have been accepted for publication and that will be published by September 2024. Authors chosen to present papers will be notified in late June 2024. All co-authors will be invited to participate in the conference. Questions about the submission process or conference planning may be addressed to [email protected]; questions about the subject matter of the meeting may be addressed to [email protected].

NBER – SI 2024 Digital Economics and Artificial Intelligence
DATE July 17-19, 2024 (US Eastern Time)
LOCATION Ballroom A, Royal Sonesta Hotel, 40 Edwin H. Land Blvd., Cambridge, MA and YouTube
ORGANIZERS Erik Brynjolfsson, Avi Goldfarb, and Catherine Tucker

Beyer, Gerry W., Artificial Intelligence Ethics for the Estate Planner (March 2024). Estate Planning Developments for Texas Professionals, Available at SSRN. Artificial intelligence (AI) is rapidly being incorporated into estate planning practices. AI products can increase the speed by which you draft, review, and summarize wills, trusts, other estate-related documents, pleadings, briefs, and client communications. Likewise, the speed by which you can conduct legal and financial research is faster than ever. AI may organize the tasks needed for an estate administration, from initial filing to final accounting. Fiduciary investment decisions may also be enhanced by the use of AI. Despite the potentially amazing benefits that await you for incorporating artificial intelligence into your estate planning practice, you may be thinking “I’d rather not” just like a potential date responded to the author’s dinner and movie invitations back in his law school days. This article starts by explaining that this type of response is not an option in today’s world; resistance to the coming of AI is futile. Borg, Wikipedia (Jan. 14, 2024) (“resistance is futile” is catch phrase for any juggernaut against which you cannot prevail). The article then turns to a detailed analysis of the ethical concerns that AI raises.

World Bank – Measuring Development 2024: AI, the Next Generation, May 2, 2024. Washington, DC. Foundational models like large language models (LLMs) have recently commanded widespread public attention—and caution—given their transformational potential for both our economy and society. Naturally, questions loom about how these AI innovations will impact the global development research and policy landscape. If used properly by the right actors, these tools might unlock enormous troves of data and create new opportunities to improve lives around the world. The World Bank’s Development Impact (DIME) department and Development Data Group (DECDG), the Center for Effective Global Action (CEGA), and and the development community at the University of Chicago are excited to explore this topic at our tenth annual Measuring Development (MeasureDev) Conference, “AI, The Next Generation.” MeasureDev 2024 will feature presentations on AI that span the measurement ecosystem: from efforts to improve and expand responsible data infrastructure in low- and middle-income countries (LMICs) and facilitate the development of a new generation of AI tools, to analysis tailoring foundational models to optimize generative AI (GenAI) including LLMs for social impact. The event will feature speakers who are shaping the way these new tools will be adopted and regulated.


Posted in: AI in Banking and Finance, Cybercrime, Economy, KM