Not everyone is keen on artificial intelligence. Here's why some businesses are skeptical.

Gorodenkoff // Shutterstock When a California doctor asked state Assemblymember Rebecca Bauer-Kahan to sign a form allowing AI to transcribe her child’s medical visit, she refused. Her concern? Intimate medical conversations are being shared with for-profit tech companies, according to an October 2024 Associated Press investigation. Her hesitation reflects a growing tension across industries as artificial intelligence transforms business operations. While AI adoption has nearly doubled in the past year, with 65% of organizations regularly using the technology, according to McKinsey’s 2024 global survey, significant sectors of the economy have reservations. Health care providers are concerned about AI “hallucinating” medical information, small businesses lack resources for AI implementation, and creative industries are fighting copyright infringement. Drata explored the hesitations that stem from a complex mix of technical limitations, ethical concerns, and regulatory uncertainties that could widen the gap between AI adopters and holdouts. Hesitations often run deeper than technical challenges. “The real barrier here is that companies realize that there needs to be an actual cultural change within the company, and they need to ask very concrete and very rigorous questions,” Kimon Drakopoulos, director of the AI for business degree program at the University of Southern California’s Marshall School of Business, told Stacker. Those questions, he said, must be specific. “Can you define what you’re trying to achieve? What are the metrics that you’re trying to achieve? And if there is a conflict within these metrics, for example, accuracy versus fairness, which one do you value more?” Different industries, a variety of concerns Nan_Got // Shuttertstock When University of Michigan researchers examined an AI transcription tool used in 40 health systems nationwide, they discovered “hallucinations”–outputs from an AI tool that are inaccurate or illogical–in 4 in 5 transcriptions reviewed, according to the AP investigation. “Nobody wants a misdiagnosis,” Alondra Nelson, former head of the White House Office of Science and Technology Policy, told the AP. The concerns are widespread–3 out of 5 Americans say they would feel uncomfortable if their health care provider relied on AI to diagnose illnesses and recommend treatments, according to a Pew Research Center poll conducted in December 2022. Similar hesitation emerges in creative industries, where copyright concerns loom large. More than a dozen copyright lawsuits about AI were filed in 2023 alone, according to a 2024 article published in Issues in Science and Technology. “For many artists and designers, this feels like an existential threat,” the authors noted. “Their work is being used to train AI systems, which can then create images and texts that replicate their artistic style.” The scale of potential infringement alarms many creators. One popular AI image generator, Stable Diffusion, trained on nearly 6 billion images “scraped from the internet indiscriminately,” including substantial copyrighted material, according to an April 2023 Harvard Business Review analysis. Companies using AI-generated content could face damages up to $150,000 per instance of willful copyright infringement. Small businesses face different challenges, particularly around expertise and resources. A 2024 report by the U.S. Chamber of Commerce Technology Engagement Center and Teneo Research found that among those businesses not yet using AI, nearly a third said one reason was that they didn’t know enough about the technology, while a quarter said they had compliance concerns. “The first question you need to ask yourself as a small or medium enterprise business is, do you have a healthy data ecosystem,” Drakopoulos said. “Do you keep track of the data? Are you tracking everything from your inventory or which customers showed up or didn’t show up for the appointment?” The talent and resource gap between large and small organizations continues to widen. While major financial firms employ hundreds of staff for AI development, smaller institutions “do not have the IT resources or expertise to develop their own AI models,” according to a 2024 analysis by the Treasury Department. Securities and Exchange Commission chair Gary Gensler warned in February 2024 about over-reliance on a few AI models, noting this concentration could lead to “herding and network interconnectedness,” problems that typically trigger financial crises. The National Institute of Standards and Technology’s 2024 risk management framework said that AI systems are vulnerable to cyberattacks, which could compromise sensitive business data. These security risks are challenging for organizations lacking sufficient IT resources, which frequently are smaller ones. The Treasury Department noted that these smaller institutions often rely on outside companies for AI tools. Larger institutions that develop tools in-house have more control over the building, testing, and monitoring of the systems to meet their specific needs and work reliably. Cautious optimism when it comes to AI