|
November 1, 2025 | Monica E. Oss
A recent survey of U.S. business leaders—AI Quarterly Pulse Survey: Q3 2025—found that 42% of those organizations are using artificial intelligence (AI), and 55% are currently piloting new uses. Not surprising.
The executives reported a range of challenges in achieving their goals with AI investments. The quality of their organizational data topped this list at 82%, up from 56% last quarter. Cybersecurity (78%, up from 68% last quarter) and data privacy (62%, down from 69%) were also key concerns. A smaller proportion of executives are concerned about regulatory uncertainty (43%, down from 55%) and lack of risk/governance processes (35%, up from 30%).
What was surprising is that these executives reported an increase in the requirements for “humans-in-the-loop” in operationalizing their AI applications (61%, up from 45% last quarter). This includes not allowing AI access to sensitive data without human oversight (63%, up from 45%) and requiring that a human validate AI outputs (42%, up from 25%). Leaders also note that they are more focused on purchasing AI from trusted tech providers (74%, up from 55%).
The need for, and cost of, “human” guidance is a key issue in the return on investment (ROI) of AI applications. In this survey, 57% expect to achieve a measurable ROI in the next 12 months.
These findings are contrary to the recent report—The GenAI Divide: State Of AI In Business 2025—which found that 95% of AI pilots across all industries are failing to achieve an ROI. Additionally, most AI pilots focus on a single pain point—with most falling short—and while these pilots work well for specific tasks, they struggle in enterprise settings because they don’t adapt to organizational workflows. The problems of AI adoption and the reasons for the lack of success in AI pilots are a combination of lack of strategic alignment, lack of implementation planning, missed integration, overlooked optimization, and lack of organizational culture change (see It’s Not The Tool).
In any event, the challenge for employers is balancing the costs of AI versus staff—and the need for different workforce skills. Not surprisingly, 83% of executives said that the AI-enabled workforce needs role-specific skills and experience, and 50% need a workforce with industry-specific skills and background.
But more problematic is the need for adaptability and continuous learning (required by 71% of organizations), critical thinking and problem-solving skills (54%), and staff with technical and programming skills (46%).
For “best practices” around AI in the health and human service field, the right strategy, the right technology, and the right set of workforce skills will all be required for maximum impact. In the year ahead, our team will continue to track the evolution of policy, practice, and investments. For more, check out our current coverage of AI in the OPEN MINDS Industry Library:
Strategic Perspectives
Executive Education
Executive Resources
News
Market Watch
And for more, join me on November 18 for the webinar, An Executive Perspective: How AI Is Shaping The Future Of Integrated Care, featuring Mike Baker, Chief Executive Officer at Heritage Health, and Michael Arevalo, Psy.D., Director of Clinical Strategy at Core Solutions.
|
