Fortune 500 Companies’ Concern Over A.I. Risks Skyrockets Nearly 500%: Study

The breakthrough of A.I. into the mainstream hasn’t just transformed productivity, workforces and society as we know it—it’s also presented new risks for the nation’s largest companies. More than half of Fortune 500 companies are pointing towards the emerging technology as a potential risk factor in their annual financial filings, according to a recent report from the research firm Arize AI.

The number of Fortune 500 companies citing A.I. risks in their SEC filings has shot up by a staggering 473.5 percent since 2022 (ChatGPT was released in November of that year), with 281 highlighting concerns about the technology compared to only 49 two years ago. “If annual reports of the Fortune 500 make one thing clear, it’s that the impact of generative A.I. is being felt across a wide array of industries—even those not yet embracing the technology,” said Arize AI’s report, which was published yesterday (August 18).

Companies within media and entertainment lead the pack in referencing A.I.’s risks, as more than 91 percent of companies in that industry have included comments about the technology in the context of risk disclosures in their most recent SEC filings. Netflix (NFLX), for example, discussed its competitive risks in its most recent annual report, noting that rivals gaining an A.I. advantage could cause its operations to “be adversely impacted.” The software and technology industry follows behind, with 86.4 percent of companies citing A.I. risks, while the telecommunications sector comes in third at 70 percent.

Besides worrying that they won’t be able to keep pace with competitors, the other primary concerns cited by the Fortune 500 include regulatory risks, security concerns and general physical and reputational harms. Potential regulation governing generative A.I. “may affect aspects of our existing business model, including revenue streams for the use of our IP and how we create our entertainment products,” said The Walt Disney Company (DIS) in its recent filing. Pharmaceutical company Vertex, meanwhile, warned that employee use of A.I. could lead to “risks relating to inappropriate disclosure of sensitive information or inaccurate information appearing in the public domain.”

Only 30 percent of A.I.-aware companies highlight its benefits

In contrast to the skyrocketing number of companies citing A.I. risks, only 30 percent of the Fortune 500 companies that specifically mentioned generative A.I. also discussed its benefits in financial filings, according to Arize AI, which described this omittance as “a missed opportunity given the potential upside to investors.”

Even some of the most dominant companies in A.I. today have used their annual SEC filings to warn about the technology. Microsoft (MSFT), in its recent annual report, said that generative A.I. features “may be susceptible to unanticipated security threats from sophisticated adversaries,” while Meta (META) warned that the company’s A.I. efforts could lead to misinformation and deepfakes, including those related to political elections.

And it isn’t just big companies that are worried about generative A.I. In the past two years, concerns over A.I. have spilled over into fields like education, where teachers have found A.I. tools are increasingly being used in schoolwork; and the art world, where artists are anxious that the technology may eventually replace them. Key A.I. figures like OpenAI’s CEO Sam Altman, meanwhile, have warned that action is needed to protect against A.I.’s potential for human extinction.

Similar waves of concern over particular digital trends have occurred in the past, according to Arize AI’s report. The research firm highlighted a 2012 emphasis on cybersecurity following a wave of data breaches the previous year at major companies like Citi, Google (GOOGL), Lockheed Martin and Sony (SONY). But even then, the share of Fortune 500 companies disclosing cyber and information security as risks rose by only 87 percent between 2010 and 2012, a figure the report described as “dwarfed” by the jump seen for A.I.