Generative AI is increasingly referenced in corporate financial statements, but not necessarily in the positive sense of intelligence and transformation. Among companies discussing the impact of generative AI, seven out of ten cite potential risks to competitive position and security, as well as the spread of misinformation.
This is the conclusion of an analysis of the annual financial reports (10-Ks) of Fortune 500 companies based in the United States as of May 1, 2024. The study compared the content of the companies’ reports with information from 2022 and looked for terms such as artificial intelligence (AI), machine learning, large language models and generative AI.
Also: How your company can best use AI: Tell your board these 4 things
More than one in five companies (22%) mentioned generative AI or large language models in their financial reports, according to analysis by technology specialist Arize. This share represents a 250% increase in the number of mentions of AI in these reports since 2022.
Public companies are required to disclose known or potential risks in their financial reports. This is one reason for the high number of not-so-positive mentions of generative AI. But the growth also illustrates the concerns raised by the new technology.
Also: Agile development can unleash the power of generative AI – here’s how
Nearly seven in 10 financial statements mentioned generative AI in the context of disclosure risks, whether through the use of new technologies or as an external competitive or security threat to the company. At least 281 companies (56%) cited AI as a potential risk factor, an increase of 474% from 2022.
Only 31% of companies that mentioned generative AI in their reports cited its benefits. Many organizations may be missing an opportunity to sell investors on AI adoption. “While many companies are likely to be overly cautious about disclosing even remote AI risks for regulatory reasons, such statements in isolation may not reflect a company’s overall vision,” the Arize authors emphasized.
The risks were not necessarily only related to bias, safety or other AI issues. The lack of timeliness in AI developments was also cited as a risk factor, as noted in S&P Global’s 10-K filing dated December 31, 2023. “Generative artificial intelligence can be used in ways that significantly increase access to publicly available free or relatively inexpensive information,” the statement said. “Public sources of free or relatively inexpensive information may reduce demand for our products and services.”
Also: It is time for companies to leave the hype about generative AI behind and find real added value
Another risk, reputational damage, was mentioned in Motorola’s 10-K filing. “As we increasingly incorporate AI, including generative AI, into our offerings, we may enable or offer solutions that are controversial due to their actual or perceived impact on social and ethical issues arising from the use of new and evolving AI in such offerings,” the financial statement said.
“AI may not always work as intended and data sets may be inadequate or contain illegal, distorted, harmful or offensive information, which could negatively impact our operating results, business reputation or customer acceptance of our AI offerings.”
Motorola said it has AI governance programs and internal technology oversight committees, but “our reputation or competitiveness could still be harmed due to inconsistencies in the application of the technology or ethical concerns, which could result in negative publicity.”
Also: When is the right time to invest in AI? 4 tips to help you decide
However, there is also positive news from the study. Generative AI was seen in a more positive light by at least a third of organizations, as Quest Diagnostics quotes in its financial reporting: “In 2023, we launched an initiative to use generative AI to improve several areas of our business, including software development, customer service, claim analysis, appointment optimization, sample processing, and marketing. We expect to further develop these projects in 2024.”
Quest also noted that the company plans to align its AI practices with the NIST AI Risk Management Framework and “enter into strategic partnerships with external AI experts as needed to ensure we stay informed of the latest technological advances in the industry.”
In an even more positive and forward-looking statement, Quest stated: “We believe that generative AI will help us innovate and grow responsibly while improving the customer and employee experience and achieving cost efficiencies. We intend to continue to be at the forefront of innovative, responsible and safe use of AI, including generative AI, in diagnostic information solutions.”
