Do android-powered project teams dream of better solutions? The answer, according to a recent Harvard Business School working paper, is yes.
In "The Cybernetic Teammate: A Field Experiment on Generative AI Reshaping Teamwork and Expertise," authors Fabrizio Dell'Acqua and colleagues (including the prolific Ethan Mollick) discuss a fascinating experiment at consumer products giant Proctor and Gamble that examines how artificial intelligence transforms collaborative work environments.
Among its most provocative insights is the finding that combining human teams with generative AI significantly increases the likelihood of producing novel ideas that push the boundaries of innovation forward, a 37% boost for individuals and a 39% boost for teams.
"These patterns indicate that AI, particularly when combined with teamwork, doesn't just improve average performance but substantially increases the likelihood of producing the kind of breakthrough solutions that drive organizational success," the authors observe.
The paper's intriguing findings emerged from a meticulously designed field experiment with 776 professionals at P&G, creating a 2x2 experimental framework that examined individuals and teams working with and without AI assistance.
Two key takeaways emerge from the research. First, human-machine collaboration powers innovation, especially when teams use AI. Second, when individuals use generative AI, they tend to envision broader solutions that avoid functional silos.
Cybernetic teammates, in a nod to computer visionary Norbert Wiener who predicted as much, unlock the creativity of humans in powerful ways.
AI Augmentation and the Top 10%
Perhaps the most striking observation of the working paper is the “breakthrough solution phenomenon,” where teams augmented with AI were three times more likely to produce solutions ranking in the top 10%.
Specifically, teams with AI demonstrated a 9.2 percentage point increase in producing exceptional solutions compared to a baseline of just 5.8% for individuals without AI, based on a scale used for all proposed ideas.
While individuals using AI showed some improvement in breakthrough potential, this effect wasn't statistically significant, suggesting that the combination of AI and teamwork creates a uniquely powerful synergy for exceptional innovation.
This finding indicates that AI expands the upper tail of possibility. In other words, AI-augmented teams aren't just reliably better on average — they're dramatically more likely to generate the kind of rare solutions that drive disproportionate value.
As the authors note, in many organizations, a handful of breakthrough ideas can generate outsized returns, an intuitive idea from innovation theory which is also echoed in the well-known work of Clayton Christensen.
What makes this finding particularly significant is how it contrasts with traditional models of innovation. Historically, breakthrough solutions often emerged from either exceptional individual talent or from carefully constructed, diverse teams.
The research suggests a third path: augmenting regular teams with AI creates a new form of "cybernetic teamwork" that can rival or exceed traditional collaboration models in generating exceptional outcomes.
Interestingly, this breakthrough potential coexists with participants' diminished confidence in their solutions. Despite objectively producing higher-quality work, AI-enabled participants were actually less likely to believe their solutions would rank in the top 10%.
Such a disconnect between actual and perceived performance suggests that organizations implementing AI must deliberately address this confidence gap, perhaps with feedback mechanisms that help workers internalize their improved capabilities.
This human-AI integration appears key to producing breakthrough solutions that balance novelty with practical application. Additionally, AI significantly reduced the time spent on tasks (approximately 12-16% less time), while simultaneously enabling the production of substantially longer and more comprehensive outputs.
This efficiency gain allowed teams to explore solution spaces more thoroughly and iteratively refine their proposals, potentially increasing the likelihood of discovering truly innovative approaches.
AI that Cross Organizational Silos
The research reveals another crucial dimension of AI's impact: its ability to break down functional silos.
Without AI, the researchers observed clear professional boundaries: Commercial specialists proposed predominantly commercial solutions while R&D professionals favored technical approaches.
By contrast, when teams worked without AI, they produced more balanced solutions through cross-functional collaboration.
Remarkably, individuals using AI achieved similar levels of solution balance on their own, effectively replicating the knowledge integration typically achieved through team collaboration.
This suggests AI serves not just as an information provider but as an effective boundary-spanning mechanism, helping professionals reason across traditional domain boundaries and approach problems more holistically.
Without AI assistance, participants tended to generate ideas closely aligned with their professional backgrounds. However, when aided by AI, this distinction largely disappeared. Both commercial and technical participants generated a more balanced mix of ideas, spanning the commercial/technical spectrum.
By leveraging AI, participants effectively expanded their problem-solving horizons, demonstrating AI's potential to foster more holistic and interdisciplinary thinking.
A similar pattern appeared at the team level. Without AI, teams exhibit a clear bimodal distribution, suggesting that solutions tend to cluster around either technical or commercial orientations, likely reflecting the dominant perspective of the more influential team member.
In contrast, AI-enabled teams show a more uniform, unimodal distribution, while maintaining similar overall levels of technical content. This shift suggests that AI helps reduce dominance effects in team collaboration, promoting more balanced contributions from both technical and commercial perspectives.
Perhaps even more significant was the finding regarding less experienced employees. When given access to AI, employees less familiar with product development achieved performance levels comparable to teams that included members with extensive experience.
This democratization of expertise has profound implications for organizational structure and team composition, potentially allowing organizations to leverage broader talent pools for innovation work.
Implications for Leadership
For organizational leaders, these findings suggest that investing in AI capabilities might be particularly valuable for innovation teams tasked with generating breakthrough solutions.
Rather than viewing AI as merely a productivity tool for routine tasks, organizations should consider its potential to expand the upper bounds of creative possibility when combined with human collaboration.
As the authors observe, generative AI is not just a fancy search engine but rather an active participant in the creation of novel ideas.
“By contributing to decision-making, creativity, and even emotional responses, AI is reshaping the conditions under which teams form and function,” the paper dryly concludes.
However, the authors also note key limitations.
The participants were relatively inexperienced with AI prompting techniques, suggesting the observed benefits represent a lower bound.
As users develop more sophisticated AI interaction strategies, the advantages of AI-enabled work could increase substantially.
Furthermore, the AI tools used were not optimized for collaborative environments. Purpose-built collaborative AI systems could potentially unlock significantly greater benefits.
Nevertheless, "The Cybernetic Teammate" is a timely piece of work that presents compelling evidence that AI integration is not merely augmenting existing work processes but reshaping the nature of collaboration and expertise in organizational settings.
The research demonstrates that AI-augmented teams can dramatically increase the likelihood of generating breakthrough solutions — the kind that drive exceptional organizational outcomes and innovation success.
As AI technologies continue to evolve, this breakthrough potential may represent their most transformative impact on organizational performance and innovation.
Agentic Foundry: AI For Real-World Results
Learn how agentic AI boosts productivity, speeds decisions and drives growth
— while always keeping you in the loop.