Emerging Use of AI and Its Relationship to Corporate Finance and Governance

Abstract

Artificial intelligence (AI) use has become a major emerging trend in corporate finance and governance. AI is used for a variety of business tasks, such as assessing credit risk, document analysis, corporate default forecasting, and detecting fraud. This study first provides an overview of the development of AI applications related to financial reporting and corporate governance and then examines the financial performance of firms rated highly for their use of AI. AI applications can improve risk management, auditing processes, financial distress, fraud detection, and board performance. The findings can help directors, managers, financial personnel, and others interested in AI.

1. Introduction

The use of artificial intelligence (AI) in business and society has become a foremost emerging trend in business innovation, operations, corporate finance, and governance. Artificial intelligence (AI) is a division of computer science that involves systems with the ability to mimic human intelligence. The primary functions of AI include analyzing data, identifying patterns in the data, problem-solving, and making predictions (Microsoft, 2025). The efficacy of AI has made its integration into business processes a focal point in the strategic innovation goals of most medium- and large-sized firms. AI facilitates the accuracy and processing speed of a number of business tasks, such as assessing credit risk, document analysis, corporate default forecasting, and detecting fraud. In this study, an overview of the development of AI applications related to financial reporting and corporate governance is presented. In addition, an examination is made of the financial performance of firms rated highly for their use of AI. This is a unique and novel contribution to the research literature; no prior study has included this current time period or evaluated this set of top AI-using firms. The results will be of interest to corporate managers, accounting and financial personnel, investors, academic researchers, and others interested in AI and its relationship to corporate finance and governance.

Many companies are using AI to varying degrees, with AI applications having been developed for use across industries (Kerr et al., 2025; Enholm et al., 2022). Regarding corporate governance, top managers ensure that a company follows established ethical practices, operates in a socially responsible way, considers the best interests of all its stakeholders, and integrates new approaches and technologies to enable the firm to succeed, providing an appropriate return on investment to its owners. The board of directors has key governance responsibilities, specifically related to the following: (1) ensure that financial reporting accurately reflects the results of the company’s operations, (2) make independent judgments about the validity and effectiveness of management’s strategic actions, (3) evaluate how well the company’s strategy is being executed, and (4) serve shareholder interests (Thompson et al., 2024).

Effective integration of AI will likely have a substantial impact on the long-term success of many corporations. Firms that develop and market AI systems, such as Anthropic and OpenAI (Brier, 2025), and firms of varied industries that make notable use of AI in their operations, such as IBM and Microsoft (L. M. Smith & Xu, 2025), are expected to have a major impact on business and society in the coming years (cf., Enholm et al., 2022).

Formally stated, this study addresses the following research question:

RQ: How does the financial performance of firms noted for extensive use of AI compare to that of other firms?

The remainder of this study includes the following sections: Review of Theory and Prior Research; Methodology, Analysis, and Results; Summary and Conclusions; and Limitations and Future Research.

2. Review of Theory and Prior Research

Artificial intelligence is fundamentally derived from information systems theory. AI systems, like other systems, consist of four parts: inputs, processes, outputs, and a feedback loop (cf. Kerr et al., 2025; Hevner & Chatterjee, 2010; Gregor, 2006). Information systems theory fundamentally explains the functioning of all types of systems, whether natural or manmade. In the information system department of every business firm and all types of organizations, these four components are continually active. In business, notably, in the accounting and financial functions, transaction information is received that is then processed to provide the financial analyses and documents necessary for the effective functioning of the organization (cf., Gregor, 2006; Bain et al., 2002).

The recent expansion in the use of AI can be explained via the economic theory of capitalism (cf., Kerr et al., 2025; Amadeo, 2024; Yolles, 2024). Stewart (2020) observes that capitalism is very different from other economic paradigms, such as Marxism, socialism, and authoritarianism; specifically, capitalism compensates for individual work ethic and inventiveness. Under capitalism, an individual can advance their economic position by working hard and developing new products or methodologies. This kind of economic opportunity is rare in human history and is not possible in numerous nations today (Stewart, 2020). In a capitalistic society, innovative companies have an incentive to respond to increasing demand for AI approaches to business, report writing, scientific applications, and search methodologies. Consumers will gravitate to companies that are meeting their needs through the creation, provision, and use of AI applications (Kerr et al., 2025; Rashid & Kausik, 2024). A competitive market under capitalism can motivate companies to operate efficiently to survive.

Some companies have deployed AI to improve marketing efforts, to assist with quality assurance and manufacturing activities, to track corporate social responsibility and sustainability outcomes, and to aid with accounting and financial processes (cf., Ibrahim et al., 2022; Kerr et al., 2025; Maali et al., 2021; Menzies et al., 2024; Enholm et al., 2022).

AI application areas include the following:

• Board performance;
• Risk management;
• Auditing;
• Financial distress management;
• Fraud detection;
• Sustainability and corporate social responsibility (CSR).

AI applications may be a component of a firm’s strategic innovation portfolio. Strategic innovation is focused on matching a company’s innovation initiatives with its strategic objectives. Rather than evaluating projects on net present value or other financial metrics, management may examine how closely innovation projects are aligned with strategic priorities (Si et al., 2023). An examination of a company’s internal and external situation has been found to disclose competitive weaknesses, market opportunities, outdated processes, new technologies yet to be adopted, and other issues of strategic importance. A strategic approach to innovation focuses on launching innovation projects, such as AI, to meet market demand and improve performance on key company metrics.

AI analytics have a tremendous impact on risk management through the technology’s ability to anticipate supply chain volatility, regulatory challenges, and labor trends (Campbell, 2025). The predictive analytics capability of AI applications is not only beneficial to management in formulating strategy but also allows managers to assess the degree to which the firm’s strategic actions are aligned with driving forces of change in the industry (Shaban & Omoush, 2025).

AI applications have been used in numerous business-related tasks, going back to early rules-based systems, commonly referred to as expert systems. These were extremely simple compared to today’s more robust Agentic AI systems, which have the capability to make autonomous decisions with little or no human involvement. Nevertheless, AI expert systems paved the way for our modern systems. For example, an early expert system pertaining to accounting for business combinations, Purpool, was developed in the 1990s (L. M. Smith & McDuffie, 1996; McDuffie et al., 1991a, 1991b). This was followed by an expert system that helped determine the appropriate type of audit report to be issued by auditors (McDuffie & Smith, 2006). Both systems are shown on the timeline of notable events for AI (Kerr et al., 2025).

In a systematic review of AI applications related to corporate governance, the use of AI deep learning (DL) applications was found to be effective in predicting financial distress and bankruptcy. DL applications have been found to outperform traditional financial analysis methods in predicting financial distress, achieving an accuracy rate as high as 93 percent (Ahdadou et al., 2024). The continuous learning nature of AI systems allows for high levels of accuracy in financial reporting despite changing financial landscapes (Shaban & Omoush, 2025). In addition, financial transparency improved by AI verification processes provides managers with up-to-date and relevant financial highlights (Odonkor et al., 2024). The availability of more timely and accurate financial data provides managers with a clearer understanding of the firm’s current financial condition and better empowers them to make decisions that serve shareholder interests. Similarly, AI applications with the capability to detect financial reporting fraud offer managers greater confidence that a company’s financial reporting accurately reflects the results of its operations (Kipp et al., 2020).

AI applications such as ChatGPT can also aid firms in ensuring compliance with governance responsibilities and tracking social responsibility initiatives and impact metrics (Ahdadou et al., 2024). AI can provide accurate and timely data supporting environmental monitoring and other social responsibility initiatives. AI has been shown to enhance data analysis efficiency and provide auditors and managers with factual data needed to make informed decisions (Nicolau, 2023). The availability of such information allows board members to assess data rather than rely on assurances of management when making judgments about how well the company’s sustainability strategy is being implemented.

The effective use of AI can be a tool supporting superior execution of internal activities and effective corporate governance. While in a nascent stage of adoption, AI offers significant benefits to boards of directors in their governance and oversight roles. AI applications in governance are progressing slowly, with less than 13 percent of directors on boards possessing knowledge and experience with AI (Campbell, 2025). AI-generated data has been shown to improve board performance (Cui et al., 2022). Decision-making by the board of directors may be improved with more transparent data, but AI can also aid in the assessment of the quality of decisions by the board. For example, board decisions have traditionally been made based upon data presented by management in the form of quarterly reports or static presentations. The availability of real-time data analytics provides an opportunity to transform oversight practices (Campbell, 2025).

The duty of the board of directors is to exercise strong oversight to see that the company’s strategies are developed and executed in a manner that benefits shareholders and meets corporate citizenship and environmental sustainability targets. Every corporation should have a strong, independent board of directors that (1) is well informed about the company’s performance; (2) guides and judges the CEO and other top executives; (3) has the courage to curb management actions it believes are inappropriate, unethical, or unduly risky; (4) certifies to shareholders that the CEO is meeting the board’s expectations; (5) provides insight and advice to management; and (6) is intensely involved in debating the pros and cons of key decisions and actions (Nadler, 2004; Gamble et al., 2024).

AI-generated data has the potential to greatly benefit audit, risk, and technology committees, compensation and human capital committees, and environmental sustainability committees (Kappagoda, 2025). In a study of 564 corporate professionals, AI-generated data was found to have a strong, significant positive relationship with the quality of corporate decision-making as reflected in improved risk management, increased transparency, higher stakeholder engagement, and effective executive compensation (Shaban & Omoush, 2025).

AI-generated scorecards can identify governance strengths and weaknesses and assess overall board performance (Cosma et al., 2018). The adoption of AI technology to assist in governance functions may benefit the performance of nominating committees by matching the subject matter expertise of members to committee assignments. Similarly, AI evaluation of board member expertise may disclose weaknesses in the makeup of the board (Kappagoda, 2025). AI assessments of board performance facilitate continuous improvement in effective governance and help ensure that board decisions serve the interests of shareholders and stakeholders.

A study by Babina et al. (2024) examined the usage and economic effects of AI technologies. The authors presented a new measure of firm-wide AI investments based on employee resumes. They found a major increase in AI investments across sectors. The findings show that AI investing led to higher sales growth, employment, and firm market valuations. Growth was attributed to expanding product innovation. Their results suggest that AI may foster growth and “superstar” firms through product innovation. They conclude the following: “Further understanding how AI affects production processes, corporate strategies, and the organizational structure of firms and assessing the distributional impacts of AI technologies across firms and workers are fruitful avenues for future research” (Babina et al., 2024).

So far, the benefits of AI have been discussed. A recent MIT study examined the use of generative AI (GenAI), which is a subset of AI. Whereas AI is primarily used for analysis and decision-making, GenAI concentrates on creating new content (Microsoft, 2025). The study looked at firms that have heavily invested in using GenAI. They found that 95% of the firms experienced no measurable effect on profit or return on investment (ROI). The researchers determined that a key problem was the firm failing to fully assimilate GenAI into business operations. While many GenAI applications were applied to marketing, the best ROI was usually obtained in streamlining business processes. Another problem was the inability of the GenAI to learn and adapt to the company’s needs. Companies that were benefiting mandated an adaptive GenAI application that could improve with feedback and integrate into business processes (Challapally et al., 2025).

Prior studies of technology-related events, such as the adoption of new technologies, technology breakdowns, cybercrime, and cyber terrorism, have discussed and/or examined outcomes, such as effects on market prices or effects on financial performance (cf., Godase, 2025; Efendi et al., 2011, 2012, 2014). In studies of cyberattacks, an event study approach examined the market reaction to news of cybercrime and cyberterrorism (K. T. Smith et al., 2011, 2023). A similar event study methodology might be implemented to measure market reactions to AI-related announcements, such as AI product launches, public AI strategy releases, etc. (cf., Getz & Page, 2024; Miller, 2023).

3. Methodology, Analysis, and Results

Our research question concerns whether the financial performance of firms noted for the extensive use of AI is different from that of other firms. We conducted two different sets of analysis to compare top AI-using firms to other firms. First, we compared top AI-using firms to their industry averages. Second, we conducted a matched-pair study, comparing top AI-using firms to other firms (cf., Bobbitt, 2021).

The top AI-using firms were obtained from lists previously provided (L. M. Smith & Xu, 2025; Bowman, 2025). Firms listed on the timeline of “Notable Events in the History of AI” include Google, Amazon, IBM, Microsoft, and Facebook [Meta Platforms] (L. M. Smith & Xu, 2025). Another list, compiled by Motley Fool, includes the following: Amazon, Meta Platforms, Tesla, Upstart, Netflix, Alphabet, JP Morgan Chase, Boeing, Johnson & Johnson, and Exxon Mobil (Bowman, 2025). Combining the two lists, 12 unique firms are identified as leading corporate users of AI, listed alphabetically as follows: Alphabet (Google), Amazon, Boeing, Exxon Mobil, IBM, Johnson & Johnson, JP Morgan Chase, Meta Platforms, Microsoft, Netflix, Tesla, and Upstart.

We gathered 2024 financial performance data (gross and net profit margin, along with return on equity) from the financial ratios suite by Wharton Research Database Services (WRDS). Market risk level (beta) was obtained from Fusion Media Limited (2025). Financial data for the top AI-using firms is shown in

Table 1. Descriptive statistics for sampled companies (fiscal year 2024).

	Top AI-Using Companies	Ticker	GICS Code	Industry Description	Revenue (Mn USD, $)	Gross Profit Margin (%)	Net Profit Margin (%)	Return on Equity (%)	Risk Level (Market Beta)
1.	Alphabet, Inc., Mountain View, CA	GOOGL	502030	Interactive Media and Services	350,018	62.1	27.7	33.0	1.01
2.	Amazon.com, Inc., Seattle, WA	AMZN	255030	Broadline Retail	637,959	56.6	8.0	23.8	1.31
3.	Boeing Company, Arlington, TX	BA	201010	Aerospace and Defense	66,517	6.6	−10.8	42.1	1.48
4.	Exxon Mobil Corporation, Spring, TX	XOM	101020	Oil, Gas, and Consumable Fuels	349,585	22.1	9.9	13.6	1.07
5.	IBM Corporation, Armonk, NY	IBM	451020	IT Services	62,753	61.4	10.2	27.1	0.73
6.	Johnson & Johnson, New Brunswick, NJ	JNJ	352020	Pharmaceuticals	88,821	77.8	16.8	20.1	0.39
7.	JPMorgan Chase & Company, New York, NY	JPM	401010	Banks	177,556	59.3	19.7	17.7	1.13
8.	Meta Platforms, Inc., Menlo Park, CA	META	502030	Interactive Media and Services	164,501	90.6	35.6	36.9	1.24
9.	Microsoft Corporation, Redmon, VA	MSFT	451030	Software	245,122	79.5	35.6	36.6	0.10
10.	Netflix, Inc., Los Gatos, CA	NFLX	502020	Entertainment	39,001	46.1	20.7	36.1	1.60
11.	Tesla, Inc., Austin, TX	TSLA	251020	Automobiles	97,690	23.5	13.3	20.9	1.78
12.	Upstart Holdings, Inc., San Mateo, CA	UPST	402020	Consumer Finance	637	−23.1	−27.8	−27.1	2.38
				Mean	190,013.3	46.88	13.24	23.40	1.18
				Median	131,095.5	57.96	15.05	25.46	1.19
				Standard Deviation	182,019.3	33.45	18.16	18.26	0.61

. The top twelve AI-using firms in our dataset had a median revenue of USD 131.1 billion in 2024, with Amazon being the largest firm by revenue at USD 638 billion in 2024 and Upstart Holdings being the smallest at USD 637 million. Our list of top AI-using firms represents some of the largest firms in the world.

Our first analysis compared our top AI-using firms to their industry groups across four measures: gross profit margin, net profit margin, return on equity, and market risk. Industry groups and their data were collected from Damodaran (2025a). Damodaran (2025a) utilizes the Standard Industrial Classification (SIC) codes instead of GICS codes as a starting basis for industry groupings. Damodaran aggregates data from several sources, including Bloomberg, Capital IQ, Compustat, and Morningstar (Damodaran, 2025b). All analyses utilized a two-sample t-test assuming equal variances. We assumed equal variances based on Levene’s test.

Table 2. Industry level of analysis for gross profit margin (%).

Company (Top AI-Using Firms)		Industry Group		Company GPM (%)	Industry GPM (%)
Alphabet, Inc.		Software (Entertainment)		62.1	65.4
Amazon.com, Inc.		Retail (General)		56.6	32.2
Boeing Company		Aerospace/Defense		6.6	17.1
Exxon Mobil Corporation		Oil/Gas (Integrated)		22.1	35.6
IBM Corporation		Computer Services		61.4	24.1
Johnson & Johnson		Drugs (Pharmaceutical)		77.8	70.3
JPMorgan Chase & Company		Bank (Money Center)		59.3	3.2
Meta Platforms, Inc.		Software (Entertainment)		90.6	65.4
Microsoft Corporation		Software (System and Application)		79.5	72.4
Netflix, Inc.		Entertainment		46.1	39.7
Tesla, Inc.		Auto and Truck		23.5	11.1
Upstart Holdings, Inc.		Financial Svcs. (Non-Bank and Insurance)		−23.1	68.4
	Mean	Std. Deviation	t-Statistic	df	p-value
Top AI-Using Firms	46.88	33.449	0.398	22	0.695
Industry Group	42.07	25.322

Note: Industry classifications and values from Damodaran (2025a). t-test assumes equal variance based on Levene’s tests for equal variances. p-value is based on a two-sided test.

shows our top AI-using firms compared to their industry groupings based on gross profit margin. The top AI-using firms had a higher mean gross profit margin compared to their industry groups, 46.88 percent compared to 42.07 percent. The difference was not statistically significant (p > 0.10).

Table 3. Industry level of analysis for net profit margin (%).

Company (Top AI-Using Firms)		Industry Group		Company NPM (%)	Industry NPM (%)
Alphabet, Inc.		Software (Entertainment)		27.7	27.4
Amazon.com, Inc.		Retail (General)		8.0	4.6
Boeing Company		Aerospace/Defense		−10.8	4.4
Exxon Mobil Corporation		Oil/Gas (Integrated)		9.9	9.8
IBM Corporation		Computer Services		10.2	4.1
Johnson & Johnson		Drugs (Pharmaceutical)		16.8	8.9
JPMorgan Chase & Company		Bank (Money Center)		19.7	25.8
Meta Platforms, Inc.		Software (Entertainment)		35.6	27.4
Microsoft Corporation		Software (System and Application)		35.6	22.9
Netflix, Inc.		Entertainment		20.7	−3.2
Tesla, Inc.		Auto and Truck		13.3	3.8
Upstart Holdings, Inc.		Financial Svcs. (Non-Bank and Insurance)		−27.8	22.3
	Mean	Std. Deviation	t-Statistic	df	p-value
Top AI-Using Firms	13.24	18.164	0.008	22	0.994
Industry Group	13.19	11.129

Note: Industry classifications and values from Damodaran (2025a). t-test assumes equal variance based on Levene’s test for equal variances. p-value is based on a two-sided test.

compares the net profit margin for the top AI-using firms and their industry groupings. The net profit margin was fairly comparable, with the AI-using firms having a mean of 13.24 and the industry groups having a mean of 13.19. The difference was statistically insignificant (p > 0.10). Summary data and analysis for gross and net profit margin are shown in Table 2 and Table 3.

Comparing return on equity (ROE), the top AI-using firms had a higher mean ROE compared to their industry groups: 23.4 percent compared to 18.54 percent. However, the difference was not statistically significant (p > 0.10). Summary data and analysis for net profit margin are shown in

Table 4. Industry level of analysis for return on equity (%).

Company (Top AI-Using Firms)		Industry Group		Company ROE (%)	Industry ROE (%)
Alphabet, Inc.		Software (Entertainment)		33.0	33.5
Amazon.com, Inc.		Retail (General)		23.8	25.4
Boeing Company		Aerospace/Defense		42.1	11.9
Exxon Mobil Corporation		Oil/Gas (Integrated)		13.6	14.2
IBM Corporation		Computer Services		27.1	17.4
Johnson & Johnson		Drugs (Pharmaceutical)		20.1	10.5
JPMorgan Chase & Company		Bank (Money Center)		17.7	11.5
Meta Platforms, Inc.		Software (Entertainment)		36.9	33.5
Microsoft Corporation		Software (System and Application)		36.6	27.7
Netflix, Inc.		Entertainment		36.1	−3.9
Tesla, Inc.		Auto and Truck		20.9	9.3
Upstart Holdings, Inc.		Financial Svcs. (Non-Bank and Insurance)		−27.1	31.5
	Mean	Std. Deviation	t-Statistic	df	p-value
Top AI-Using Firms	23.40	18.257	0.776	22	0.446
Industry Group	18.54	11.722

Note: Industry classifications and values from Damodaran (2025a). t-test assumes equal variance based on Levene’s test for equal variances. p-value is based on a two-sided test.

. In the most recent industry analysis concerning risk level (Beta), the top AI-using firms had a slightly higher risk level compared to their industry groups: 1.18 compared to 1.08. The difference was statistically insignificant (p > 0.10). Summary data and analysis for net profit margin are shown in

Table 5. Industry level of analysis for market beta (β).

Company (Top AI-Using Firms)		Industry Group		Company Beta (β)	Industry Beta (β)
Alphabet, Inc.		Software (Entertainment)		1.01	1.18
Amazon.com, Inc.		Retail (General)		1.31	1.06
Boeing Company		Aerospace/Defense		1.48	0.90
Exxon Mobil Corporation		Oil/Gas (Integrated)		1.07	0.48
IBM Corporation		Computer Services		0.73	1.23
Johnson & Johnson		Drugs (Pharmaceutical)		0.39	1.07
JPMorgan Chase & Company		Bank (Money Center)		1.13	0.88
Meta Platforms, Inc.		Software (Entertainment)		1.24	1.18
Microsoft Corporation		Software (System and Application)		0.10	1.24
Netflix, Inc.		Entertainment		1.60	1.04
Tesla, Inc.		Auto and Truck		1.78	1.62
Upstart Holdings, Inc.		Financial Svcs. (Non-Bank and Insurance)		2.38	1.07
	Mean	Std. Deviation	t-Statistic	df	p-value
Top AI-Using Firms	1.18	0.613	0.539	22	0.595
Industry Group	1.08	0.268

Note: Industry classifications and values from Damodaran (2025a). t-test assumes equal variance based on Levene’s test for equal variances. p-value is based on a two-sided test.

.

The second analysis compared our twelve AI-using firms to a set of twelve matching companies. We used the Mergent Intellect by FTSE Russell database to identify competitive firms to compare against the top AI-using firms. Once a list of competitive firms was identified, we then used Compustat’s North America Fundamentals database to ensure that competitive firms were within the same industry. Companies were considered to be operating in the same industry if their Global Industry Classification Standard (GICS) industry codes matched. We gathered 2024 financial performance data (gross and net profit margin and return on equity) from the financial ratios suite by Wharton Research Database Services (WRDS). Market risk level (beta) was obtained from Fusion Media Limited (2025). Our matched firms had a median revenue of USD 13.4 billion in 2024. While much lower than the revenue of our top AI-using firms (USD 131 billion), their revenue places the matched firms in the top 0.02 percent of businesses in the United States by conservative estimates. The descriptive statistics for the twelve matched companies are shown in

Table 6. Descriptive statistics for matched companies (fiscal year 2024).

	Matched Company	Ticker	GICS Code	Industry Description	Revenue (Mn USD, $)	Gross Profit Margin (%)	Net Profit Margin (%)	Return on Equity (%)	Risk Level (Market Beta)
1.	Snap, Inc., Santa Monica, CA	SNAP	502030	Interactive Media and Services	5361	59.3	19.7	17.7	0.62
2.	Coupang, Inc., Seattle, WA	CPNG	255030	Broadline Retail	30,268	49.6	12.4	9.0	1.16
3.	Northrop Grumman Corporation, West Falls Church, VA	NOC	201010	Aerospace and Defense	41,033	90.6	35.6	36.9	0.13
4.	Chevron Corporation, Houston, TX	CVX	101020	Oil, Gas, and Consumable Fuels	202,792	79.5	6.2	7.1	0.91
5.	GoDaddy Inc., Tempe, AZ	GDDY	451020	IT Services	4573	57.03	41.34	343.57	1.02
6.	Merck & Co., Inc., Rahway, NJ	MRK	352020	Pharmaceuticals	64,168	92.0	25.6	34.2	0.38
7.	Bank Of America Corporation, Charlotte, NC	BAC	401010	Banks	192,434	46.1	20.7	36.1	1.35
8.	Pinterest, Inc., San Francisco, CA	PINS	502030	Interactive Media and Services	3646	39.6	8.5	15.2	1.16
9.	Adobe Inc., San Jose, CA	ADBE	451030	Software	21,505	23.5	13.3	20.9	1.49
10.	Roku, Inc, San Jose, CA	ROKU	502020	Entertainment	4113	46.51	−4.42	−7.34	2.05
11.	Rivian Automotive, Inc., Irvine, CA	RIVN	251020	Automobiles	4970	−23.1	−27.8	−27.1	1.81
12.	SoFi Technologies Inc., San Francisco, CA	SOFI	402020	Consumer Finance	3766	68.80	6.03	0.04	1.94
				Mean	48,219.1	52.45	13.09	40.52	1.17
				Median	13,433.2	53.33	12.82	16.48	1.16
				Standard Deviation	72,333.2	31.45	18.23	97.27	0.60

Note: Revenue collected from Mergent Intellect by FTSE Intellect. Gross profit margin, net profit margin, and return on equity collected from the financial ratios suite by Wharton Research Database Services (WRDS). Risk level collected from Fusion Media Limited.

.

t-tests were used to compare the means of top AI-using firms and their matched firms across four different metrics. We first tested the assumption of homogenous variance using Levene’s test. For each of our comparisons, there was insufficient evidence to reject the assumption of equal variance. As a result, we used a two-sample t-test with the assumption of equal variance for each of our tests. The results of the analysis are shown in

Table 7. t-test analysis for AI-using companies and matched companies.

Variable		Top AI-Using Companies	Matched Companies	t-Statistic	p-Value
Gross Profit Margin (%)	Mean	46.88	52.45	−0.420	0.679
	Variance	1118.815	989.109
	Observations	12	12
	df	22
Net Profit Margin (%)	Mean	13.24	13.09	0.021	0.984
	Variance	329.930	332.494
	Observations	12	12
	df	22
Return on Equity (%)	Mean	23.40	40.52	−0.599	0.555
	Variance	333.317	9461.361
	Observations	12	12
	df	22
Risk Level (Market Beta)	Mean	1.18	1.17	0.066	0.948
	Variance	0.375	0.365
	Observations	12	12
	df	22

Note: All t-tests assume equal variance based on Levene’s test for equal variances. p-value is based on a two-sided test.

.

The top AI-using firms had a gross profit margin of 46.88 percent in 2024, while the matched firms had a gross profit margin of 52.45 percent. While it appears that the matched firms had a higher gross profit margin, the difference was statistically insignificant (p > 0.10). Similarly, the top AI-using firms had a net profit margin of 13.24 percent in 2024 compared to 13.09 percent for the matched firms, but the difference was statistically insignificant (p > 0.10).

Top AI-using firms had a return on equity of 23.40 percent compared to the matched firms that had a return on equity of 40.52 percent; however, the difference was not statistically significant (p > 0.10). Finally, our top AI-using firms had a market beta of 1.18, while the matched firms had a market beta of 1.17. This difference was also statistically insignificant (p > 0.10). A matched-pair analysis failed to find significant differences between top AI-using firms and their counterparts. Given that the ex-post selection approach of top AI-using firms likely biases the data toward finding positive results, the failure to reject the null hypothesis provides even stronger evidence that, at least for now, AI adoption is not related to significant positive associations.

4. Summary and Conclusions

Businesses and other organizations indicate the increasing use of artificial intelligence (AI) in their operations, which are connected to financial reporting and corporate governance-related activities. Specific applications have been developed, such as enhancing marketing efforts, assisting with manufacturing and quality assurance activities, tracking corporate sustainability outcomes, and aiding with accounting and financial processes. Several AI application areas have been identified that can bolster corporate governance and financial reporting, notably risk management, auditing processes, managing financial distress, fraud detection efforts, and corporate social responsibility efforts, including sustainability.

An examination of the financial performance and risk level of firms identified as top users of AI shows that these top AI users were not significantly different from other firms. Even though AI-using companies had a higher gross profit margin and return on equity (ROE) than the industry average, the difference was not statistically significant. Our finding supports the MIT study that showed 95% of firms heavily invested in GenAI have not experienced an increase in profit or return on investment. Thus, while AI has great potential, significant outcomes on firm performance have yet to be realized. This suggests that the benefits of AI use, at least for this sample of firms, may be more limited than some speculate. However, AI use is a fairly recent phenomenon, and the impact could become significant in the future. Currently, the percentage of board directors with knowledge and experience of AI is low.

AI can enhance corporate governance, including improved risk management, heightened fraud detection, predicting financial hardship, and strengthened oversight by the board of directors. These findings should be of interest to corporate managers, accounting and financial personnel, investors, academic researchers, and others interested in AI and its relationship to corporate finance and governance.

5. Limitations and Future Research

This study is limited by the prior research examined. While this literature review was considerable, future studies might extend the literature review to incorporate additional past studies and subsequent research. This study is limited by the sample of 12 top AI-using firms; future studies can include additional firms, if and when they are identified as top AI users. This study is limited by the statistical methodologies used; future studies might include other methodologies. This study is limited by the time period examined. Subsequent studies could investigate additional time periods. A future study could use an events study methodology to measure market reactions to AI-related announcements (e.g., public AI strategy releases and AI product launches). In addition, if data can be obtained, a future study might consider using an events approach to measure the financial performance impact of AI adoption, comparing pre- and post-AI adoption on a sample of firms. Further, the adoption of AI applications is likely subject to significant lag effects. If pre- and post-adoption data can be obtained, a future study could seek to estimate the expected time horizon for AI-driven performance changes. Finally, this study was limited by the variables included in the analysis. Future studies can incorporate other financial performance variables.