According to CNBC, researchers from Harvard Business School’s Digital Data Design Institute (D^3), led by chief operator Jen Stave, partnered with Procter & Gamble to study real-world AI use in white-collar work. Their key finding is that individuals equipped with AI can perform at levels comparable to entire teams that don’t have AI access, effectively reproducing some benefits of human collaboration. The research indicates that while AI boosts individual speed and performance, the highest-quality and most innovative outcomes actually come from strategically curated, AI-enabled teams. Stave argues that mature companies are now thinking through changes in roles, aiming to protect and even add human jobs where they provide a competitive advantage, rather than simply replacing people. Furthermore, the D^3 experiment showed that AI integration can significantly reduce knowledge gaps between different pockets of domain expertise within an organization.
The collaboration paradox
Here’s the thing that’s really fascinating about this Harvard study. We’ve been told for years that collaboration is the golden ticket to innovation, right? More brains, more ideas. But now, a single person with a capable AI can apparently match the output of a whole team. That’s a mind-bending shift. It basically means AI is acting as a force multiplier for individual cognition, compressing what used to require group discussion and synthesis into a solo activity. But—and this is a huge but—the study found that the *best* work, the truly innovative solutions, still emerged from teams that were built with AI in mind. So it’s not about the AI replacing the team. It’s about the team using AI to become something new entirely.
Augmentation over automation
This gets to the core of what a lot of companies are getting wrong. The easy, knee-jerk reaction is to see AI as a pure automation tool: “This task used to take five people, now the AI can do it, so we need four fewer people.” The Harvard research, and Stave’s comments, suggest that’s a immature, and probably less effective, approach. The mature mindset she talks about is asking a harder question: “Now that one person with AI can do the work of five, how do we redesign our teams and roles to leverage that superpower for even greater outcomes?” That might mean protecting human roles that involve creativity, judgment, and interpersonal skills—the areas where, for now, we still have the upper hand. It’s about using AI to elevate human work, not just erase it.
Bridging the knowledge gaps
One of the more practical findings from the P&G experiment is AI’s role as a knowledge democratizer. In any big company, you have silos. The engineering team has deep technical knowledge the marketing team doesn’t. HR has policies R&D might not fully grasp. An AI with access to a centralized knowledge base can act as a bridge, making any team’s output more informed and universally applicable. This is huge. It means AI isn’t just making individuals faster; it’s making the *organization* smarter and more cohesive by breaking down internal barriers. The potential here is for AI to create a common operational language, which is a massive unlock for efficiency and innovation that goes way beyond simple task completion.
The future is cybernetic
Look, the researchers used tools not even optimized for collaboration. Think about that. We’re seeing these effects with general-purpose chatbots and assistants. Now imagine AI systems purpose-built to be teammates—systems designed to debate, brainstorm, and iterate *with* a human group. The study hints that this is where the next wave of productivity will come from. The goal isn’t a lonely employee staring at a chatbot. It’s a hybrid, cybernetic team where human and artificial intelligence are woven together, each playing to their strengths. Companies that figure out this new organizational design first will have a serious edge. The rest will just be saving a few minutes on drafting emails.
