Artificial Intelligence (AI) is likely to transform workplaces, fundamentally changing tasks, teamwork, and organizational dynamics. Several recent studies highlight promising early findings about how AI is impacting knowledge workers, suggesting it can actively collaborate with humans, enhancing interactions and capabilities across various professions.
The first study, conducted by a team of researchers from Harvard, Wharton, and Procter & Gamble (P&G), involving 776 experienced professionals at P&G, explored how AI (GPT-4 or GPT-4o) influences the innovation process. Participants were randomly assigned to four experimental conditions: individuals without AI, teams of two humans without AI, individuals with AI assistance, and teams of two humans with AI support. Each group tackled realistic business challenges, closely replicating tasks and scenarios they encountered in their everyday roles at P&G.
The results were remarkable: teams solely comprised of humans outperformed individuals working alone, yet individuals equipped with AI matched the performance levels of human teams. Teams using AI were 9.2 percentage points more likely to deliver top 10 percent solutions—nearly three times more effective than teams without AI. Both AI-enhanced groups operated more efficiently, completing tasks 12–16 percent faster and producing solutions that were more detailed and comprehensive.
A more important finding is that AI significantly impacted how team expertise is leveraged in new product development tasks, particularly benefiting employees who are less familiar with such tasks. Additionally, AI transforms team collaboration dynamics, reducing the traditional divide between commercial and technical employees. While employees’ previously proposed ideas aligned with their specialties, AI assistance enabled both groups to generate more balanced, interdisciplinary ideas without compromising solution quality.
This finding is particularly intriguing because it offers evidence for why this wave of AI-democratizing access to expertise differs from past technology waves. Historically, transformative technologies—from the printing press to computers and eventually the Internet—primarily democratized access to information, enabling more people to acquire knowledge previously reserved for the few. However, today’s AI democratizes expertise itself, enabling vast amounts of specialized knowledge to be applied directly to complex, specialized tasks. In the past, the only way to access more expertise was to provide more training to workers, hire higher-skilled workers, or engage specialized consultants. AI now provides an additional option, enabling a broader spectrum of people to perform tasks once reserved for experts. This is why NVIDIA CEO Jensen Huang believes that “The IT department of every company is going to be the HR department of AI agents in the future.”
The second study, published by the Commonwealth of Pennsylvania, documented early findings from their ChatGPT Pilot program, which involved more than 175 government employees exploring AI implementation across state operations. Nearly 48 percent of participants had never used ChatGPT before this pilot program, yet 85 percent reported positive experiences, estimating they saved an average of 95 minutes daily. Key tasks included writing assistance and drafting emails (36 percent), researching and exploring new topics (27 percent), summarizing documents (13 percent), and technical tasks such as coding and spreadsheet management (eight percent). One significant achievement was using ChatGPT to consolidate 93 IT policies into 34 streamlined versions.
These real-world studies underscore AI’s value as a collaborative partner, a sentiment expressed by one of the Commonwealth participants who described the tool as an “incredibly reliable and resourceful counterpart.” It echoes the findings of a paper Google released on its Co-Scientist where researchers found “how collaborative and human-centered AI systems might be able to augment human ingenuity and accelerate scientific discovery.”
They also reflect some recent findings from the AI for Workers & Learners survey conducted by Jobs for the Future (JFF), which surveyed over 2,700 individuals across various demographics and employment backgrounds. The JFF survey highlights that 57 percent of workers already experience AI significantly impacting their jobs, primarily through reducing workloads, automating routine tasks, and shifting responsibilities toward more creative and strategic endeavors. More than 77 percent of respondents said they believe AI will impact their job or career in the next 3–5 years.
However, the survey also underscores existing gaps and challenges, particularly in formal training and equitable access to AI resources. Only 31 percent of respondents indicated receiving employer-provided AI training, despite a clear interest and demand for such training. Addressing these training gaps is crucial as organizations aim to leverage AI’s transformative potential fully.
Although these studies are early explorations into AI’s role in the workplace, their findings are encouraging. AI tools like ChatGPT are poised to democratize access to expertise, breaking down traditional barriers and bringing specialized expertise directly to teams. As organizations embrace this shift, they have an exciting opportunity—not just to improve efficiency but to fundamentally transform how teams interact, innovate, and thrive in an increasingly AI-driven workplace.
The post Generative AI: The Emerging Co-Worker Transforming Teams appeared first on American Enterprise Institute – AEI.
