Before We Measure AI’s ROI, Let’s Decide What We Want It To CreateThe Female Quotient Before we decide how to measure the ROI of enterprise AI, we should decide what we want AI to create. We are still early, which means the assumptions leaders make now will shape how companies design, deploy, and measure AI for years to come. If we enter this new era with an old definition of productivity, we will measure hours saved, tasks automated, overhead reduced, and jobs eliminated.While those measures tell us what AI removes, they do not tell us what AI makes possible. If AI allows a company to do the same work with fewer people, that is efficiency. If it allows the company to create new products, enter new markets, solve new problems, and move people into higher-value roles, that is transformation.I call this Return on Intelligence: the new growth, capabilities, and human potential a company creates by combining artificial and human intelligence.Measuring What AI CreatesTraditional ROI tells us whether a technology paid for itself. Return on Intelligence asks whether the organization became more capable, competitive, and valuable because of it. That entails measuring more than cost savings. Leaders should look at the revenue AI creates through new products, services, and markets. They should measure whether AI accelerates innovation, strengthens customer relationships, improves decisions, builds intellectual property, and increases the organization’s ability to anticipate and respond to change.Companies should also measure expanded human capacity. When AI removes repetitive work, what are employees able to do with the time it releases? Are they solving more important problems, developing ideas, building relationships, exercising judgment, and creating new sources of value?MORE FOR YOUSaving ten hours has limited value if those hours simply disappear from a report. The real return comes from reinvesting that capacity in higher-value work.The AI Gender Gap Can Become a Leadership OpportunityHow companies define AI’s return will also determine how they approach their people. If leaders define success primarily through headcount reduction, the workforce conversation will focus on which jobs AI can eliminate. If they define success through growth, the conversation becomes what new businesses AI can help build and what new talent those businesses will require.This distinction is especially important for women. New LinkedIn research found that women accounted for only 26% of new hires into AI roles in the U.S. in 2025, compared with approximately 50% of new hires in non-AI occupations. Across AI companies in 27 countries, women hold just 13% of C-suite AI leadership roles.At the same time, the International Labour Organization reports that female-dominated occupations are almost twice as likely as male-dominated occupations to be exposed to generative AI. Women risk being more exposed to the jobs AI changes and less represented in the jobs AI creates.But the AI gender gap does not have to leave women behind. If we act intentionally, AI can catapult women forward and dismantle the old hierarchies that have kept leadership looking the same for generations.We Cannot Automate the Bottom Rung and Leave the Ceiling IntactAI is creating new roles, industries, and definitions of leadership. Many of the most consequential AI positions barely existed a few years ago, which means the pathways into them are not yet fixed.That gives us a once-in-a-generation opportunity to put women at the top from the beginning, as architects of how AI is designed, applied, governed, and used to close gaps in healthcare, caregiving, access to capital, financial security, advancement, education, and opportunity.We cannot build the workplace of the future on the power structures of the past. We cannot automate the bottom rung of the career ladder while leaving the ceiling intact. And we cannot retrofit women into yesterday’s workplace or ask them to keep climbing a ladder built in another era.This is our moment to redesign the entire ladder. Some of the strongest future AI leaders may not have “AI” in their job titles today. They may work in marketing, finance, operations, healthcare, customer experience, legal, communications, or human resources. They understand the business, know the customer, recognize unmet needs, and can connect technology to human problems.Companies should hire for skills and potential, not only traditional credentials. They should give women access to AI training, stretch assignments, transformation teams, budgets, sponsorship, and decision-making authority. They should invest in women building AI companies, and provide access to the customers and procurement opportunities those businesses need to scale.Most importantly, women should not simply be taught how to use AI systems after they have already been designed. Women should be architects of how AI is used to solve the problems they understand deeply.Design the Return We WantEvery enterprise AI initiative should begin with a broader set of questions. What new value can this create? What customer problem can we solve? What decision can we improve? What organizational capability can we build? What human capacity can we expand? Which employees can grow into the new roles this technology creates?The companies that win with AI will not be the ones that eliminate the most. They will be the ones that create the most new value and expand what their people are capable of achieving. The return on AI is not predetermined. It will reflect the outcomes leaders choose to design, fund, and measure. The question is not how women can catch up in AI; it is how AI can help catapult women forward. The future of work is being built right now, and women should not have to find their place in it; they should be building it. Before we calculate AI’s return, let’s design the return we want.