AI adoption isn't a technology problem; it's a management problem. That's the surprising consensus from more than 300 corporate executives, economists, venture capitalists and researchers who gathered at MIT's Business Implications of Generative AI (BIG.AI@MIT) conference in early April to discuss where AI creates value and why so many investments aren't paying off yet .

Why Are Companies Struggling to Get AI Right?

The conference's opening panel tackled one of today's biggest misconceptions: that AI adoption is primarily about selecting the right technology platforms or tools. In reality, the challenge is far more fundamental. Jim Wilson, Global Managing Director of Technology Research at Accenture, outlined a management playbook he's observed working across industries.

"Each of those five principles is a human-led activity. This is not simply about uploading a new version of Claude into your company's systems kind of passively. Active human involvement, human agency, asking feedback from workers and leadership taking a stake in this is really critical," said Wilson.

Jim Wilson, Global Managing Director of Technology Research, Accenture

Julia Neagu, an AI researcher at Databricks following the acquisition of her company Quotient AI, echoed this perspective from a builder's viewpoint. She noted that while AI has shown the most success in coding tasks, other use cases require significantly more organizational lift.

"There's definitely an expectation that AI works like magic. They can just onboard it within your organization or among your teams and it will just work. And that's just not how things happen in practice," explained Neagu.

Julia Neagu, AI Researcher, Databricks

What Does a Successful AI Adoption Playbook Actually Look Like?

Rather than focusing on which AI tool to purchase, business leaders should be asking whether their organization is structured to adopt AI effectively. According to the Accenture executive, companies that are driving real results from their AI systems follow a consistent framework :

• Process Redesign First: Start by rethinking how work gets done, not just automating existing processes as they are.
• Human-Centered Experiments: Run small-scale tests that keep people in the loop and gather feedback from actual workers.
• Governance Investment: Build clear policies and oversight mechanisms to guide how AI is used across the organization.
• Data Infrastructure: Establish a solid foundation of clean, organized data that AI systems can actually work with.
• Human Skills Development: Invest as much or more in training people as you do in the technology itself.

The insight here is critical: each of these five elements is fundamentally a human-led activity, not a technical one. The companies getting results aren't those that passively deploy new AI tools; they're the ones where leadership actively shapes how AI integrates into their operations.

Why Aren't AI Investments Showing Returns Yet?

Many business leaders are feeling pressure from their boards to demonstrate near-term returns on AI investments. The conference introduced a useful framework for understanding why that pressure may be premature: the J-curve .

The J-curve illustrates that companies investing in AI are currently in a temporary productivity dip. This isn't because AI isn't working; it's because the organizational transformation required to unlock AI's value takes time, resources and effort that don't show up immediately in output metrics. In other words, productivity dips before it rises.

Ernie Tedeschi, Chief Economist at Stripe, added important historical context. Many people incorrectly treat the period since ChatGPT's launch in late 2022 as the start of AI's economic impact. In reality, we're still in the very early stages of adoption.

"That's not even the first scene of AI. That's like the orchestra warming up before the overture," noted Tedeschi.

Ernie Tedeschi, Chief Economist, Stripe

This framing gives leaders a way to have a more honest conversation with their boards about realistic timelines for AI ROI (return on investment). The pressure to show immediate returns may actually undermine the deeper organizational changes needed to make AI truly valuable.

What Skills Will Actually Matter in an AI-Driven Workplace?

A surprising theme emerged during the conference's final panel discussion: as AI takes on more routine tasks, distinctly human skills become more valuable, not less. Creativity, judgment, accountability and the ability to connect with other people aren't soft skills on the margins anymore. In a world where AI handles execution, they become the core work itself .

Laura Burkhauser, CEO of Descript, explained how this shift is already changing hiring and organizational structure. As AI agents replace junior roles built around executing orders, companies need people who can serve as the AI agent's boss.

"We look for people who can think like an engineer or designer. Someone who's able to hold a lot of context and translate that context to an agent, oversee the way the agent is using that context and drive it when it's going the wrong way," said Burkhauser.

Laura Burkhauser, CEO, Descript

This represents a fundamental shift in how organizations should think about workforce development. Rather than asking "What can AI do?" companies should also be asking "What kind of people and structures do we need to make the most of AI?" The answer isn't fewer skilled workers; it's workers with different, more strategic capabilities.

The BIG.AI@MIT conference made clear that the next phase of AI adoption will be decided not by technology choices, but by how well organizations redesign their processes, invest in their people and maintain human oversight of AI systems. Companies that treat AI adoption as primarily a management challenge, rather than a technology one, are the ones most likely to see real returns on their investments.