Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

Unlock scale with inclusive AI

AI will only deliver competitive advantage when it actually works for people and society. Here's what leaders need to know.

Unlock scale with inclusive AI

[Adobe Stock / Harsha]

Profile photo of Rebecca Finlay

CEO · Partnership on AI

Are you excited to serve more customers in more markets by expanding your successful AI pilot across your organization? You are not alone. I hear this from many organizations. However, too often today, enterprise-wide AI deployments are losing momentum as costs increase and the pace of adoption decelerates. The slow down is impacting companies’ competitive advantage. 

Today’s AI landscape is a study in contrasts. On one hand, AI products are predicted to present a significant economic opportunity. One report cites USD $1.5 trillion as the low estimate for projected AI revenue by 2040. Another shows how AI is already embedded in the technology of our daily lives. 

Yet, reports also show that companies are experiencing significant challenges as they try to scale from pilots to org-wide deployment. According to an S&P survey, the share of companies winding down their AI implementations rose to 42% in 2025. Gartner predicts that over 40% of agentic AI projects will be canceled by the end of 2027.

Scaling up pilots with AI cannot be a “one size fits all” approach. The power of algorithmically driven systems to efficiently customize products and services for one group will falter when applied to another. To succeed in scaling AI, take into account the perspectives of all the people impacted by AI in your company, including employees, customers, and the communities you serve. 

Amazon’s early attempt to apply AI to recruiting the most promising candidates makes this clear. Having trained their AI on the resumes of their own—mostly male—top talent, they couldn’t remove the AI systems’ bias against women. No matter what the team did to remove gender bias from the training data, the system continued to rank resumes with female-sounding characteristics lower.

Your customers know this already. A 2024 Pew survey reports that both experts and the public believe that the perspectives of both men and white adults are better represented in AI design than those of women as well as of Asian, Black, or Latino adults. 

By consulting a broad group of stakeholders up front, you will save time and money over the long-term, while boosting your brand and protecting against reputational risk. 

Doing so will also improve products and services. For example, closed captioning not only ensures that people with hearing difficulties have access to spoken content, but serves as a convenience for those who may be unable to access reliable audio (or have simply run out of power on their headphones while sitting in a crowded cafe). 

GET UP TO SPEED ON PUBLIC ENGAGEMENT

Begin by learning the difference between user testing and public engagement. User testing is necessary to assess product and service function, but is not sufficient for scaling up. To ensure that all potential issues and opportunities are identified and addressed, participatory public engagement seeks out the advice of groups directly and indirectly affected by the use of AI in your products and services.

INVEST IN THE RIGHT PEOPLE

Build a great AI development team that includes experts who understand how to navigate complexity and can build respectful interactions with diverse audiences. Make sure to prioritize this work alongside that of technologists and engineers, allowing for sufficient time to capture and embed what you learn as you build. 

BE A LEADER IN INCLUSIVE AI 

Join leaders from some of the most successful companies globally who are working with respected experts in academia and civil society to develop best practices in scaling up AI. At organizations such as my own, Partnership on AI, we know that AI will only deliver competitive advantage when it actually works for people and society. 

In 2019, 181 CEOs of major corporations—including big tech firms—affirmed that the purpose of a corporation is to create value for all stakeholders—not just shareholders. This commitment is as crucial today as it was six years ago, despite the reported winding down of corporate initiatives to engage with underrepresented communities. 

Developing AI inclusivity is not a “check the box” activity. Done well, it allows companies to overcome their scaling up challenges with better, more timeless products and services that consistently meet the expectations and serve the needs of their customers and communities.