Wharton professor says AI adoption is stuck on organizational change
Eric Bradlow says companies need to redesign work, training and accountability before AI can deliver broad business gains.
By Hana Yoshida · Markets Reporter
3 min read
Many companies are still struggling to turn AI experiments into broad changes in how work gets done, according to Wharton professor Eric Bradlow. Bradlow told Fortune that the main obstacle is organizational change, including how leaders keep people involved as AI becomes part of daily decision-making.
Bradlow, Wharton’s vice dean of AI and analytics, described AI as the most consequential innovation of his lifetime in a Wharton faculty video series on American business innovation. He said people are likely to see benefits from artificial intelligence before companies complete the larger transformations many executives expect.
In an interview with Fortune, Bradlow said the technology itself is not the chief barrier. He said organizations still have not figured out how to add AI across the business in a coordinated way, and that humans remain essential to the process.
Bradlow is a computer scientist and statistician who has worked at Wharton for 30 years, according to Fortune. He has led a data science program for 20 years and has focused on AI for the past decade.
Deep skills still matter
Bradlow also worked on the Wharton-Accenture Skills Index, which Fortune said tracks more than 150 million unique U.S. profiles and 100 million job postings. He has argued that large language models increase the value of deep expertise because skilled people are needed to train AI systems and judge whether their outputs are right.
That view runs against fears that AI adoption will mainly reduce headcount. In Wharton’s video series, Bradlow said stronger companies will move people into higher-value work rather than focus on eliminating employees.
Asked by Fortune about companies that see AI integration as a broader redesign of work that could require fewer employees, Bradlow said the largest opportunities are tied to revenue growth rather than lower costs. He said companies will use AI to build new business models, creating a need to move talent across current and new roles.
That shift puts pressure on companies to train workers for new responsibilities. Fortune reported that training and reskilling are a major concern for finance chiefs as companies try to move AI beyond narrow pilots.
Governance is another gap
A KPMG report released this week pointed to a related management problem. KPMG’s Global AI Pulse Q2 2026 report, based on a survey of about 2,100 senior leaders in 20 countries, found that 75% said their CEO actively treats AI as a strategic priority.
Accountability was less settled, according to KPMG. The firm found that 24% of organizations identified the CEO or executive committee as ultimately responsible for AI-informed decisions, while 29% assigned that responsibility to a named C-suite executive.
KPMG also found that only 35% of organizations said they have very clear guidance on when people should override AI outputs. Companies with clear executive accountability were more likely to strongly agree that they can future-proof their AI strategy, at 60% compared with 22% for others, according to the report.
Bradlow’s argument, as reported by Fortune, is that leaders must put as much effort into skills, governance and accountability as they put into AI tools and infrastructure. Without that work, companies risk keeping AI confined to limited uses instead of applying it to how the business grows.
This story draws on original reporting from Fortune.