Beyond the hype: 8 drivers for true AI transformation in the agentic age

The true winners of the AI era will be businesses that master both innovation and operations.

As firms and organizations across sectors dive headfirst into their transformation journeys, a significant gap persists between artificial intelligence (AI) incubation and adoption in production. To come out ahead, leaders will need to bridge this divide, translating AI's immense capabilities into real world benefits for employees, customers and society.

At Rakuten, we've formalized our approach to AI transformation through what we call "AI-nization." AI-nization is about more than implementing AI tools or automating tasks; it's a fundamental cultural and operational shift, systematically embedding trustworthy and useful AI into every facet of our global operations, product development and planning.

To guide this AI transformation and ensure our AI initiatives are intelligent, efficient, ethical and productive, we developed a framework of “Eight Drivers of AI-nization.” These drivers provide a structured, actionable roadmap that I believe can help any organization thrive in the agentic era:

Too many companies today allow the “fear of missing out” to guide their AI strategies, resulting in costly, disjointed AI programmes.

An MIT study from earlier this year found that 95% of GenAI pilot projects fail to deliver measurable business impact, resulting in a concerning gap between AI investment and returns.

At Rakuten, we are guided by our vision of augmenting creativity with the power of AI. To achieve this goal, top management has laid out a clear vision – one that sees around the corner.

AI-nization combines leveraging our unique data assets and extensive channels to foster growth across the Rakuten Ecosystem, continuously strengthening our deep learning foundations and investing in large language models.

AI transformation requires specific, quantifiable targets that span divisions horizontally and teams vertically.

Our "Triple 20" initiative, which set the target of utilizing AI to achieve a 20% efficiency gain in marketing spend, a 20% improvement in operational processes and a 20% efficiency gain for our merchant and business partners, exemplifies this. Each BU, division and team can set their own KPIs to align with these North Star metrics.

While our future KPIs are sure to be even more ambitious, these metrics, set in 2024, helped ensure that AI was directly contributing to business growth and efficiency at a time when many organizations were still in an exploratory stage.

As GenAI's reasoning and conversational capabilities become increasingly powerful, novel risks arise. Our commitment to safety and ethical governance must keep pace.

This driver underscores the importance of pursuing innovation while remaining grounded and vigilant to maintain trust. It requires establishing robust governance structures and AI codes of ethics, ensuring continuous compliance and engineering solutions that embed safety and transparency into AI systems.

At the core of our approach to AI is the concept of omotenashi: the uniquely Japanese approach to thoughtful hospitality that exceeds expectations. We want to leverage AI not only to meet customer needs but to mindfully anticipate what they want.

Our AI-powered shopping assistant on Rakuten Ichiba, for instance, streamlines decision-making, embodying omotenashi in the digital realm. Effective AI transformation is inherently customer-centric; if the result is not a better experience for your users or stakeholders, you’re doing it wrong.

Technology fundamentals matter. You can’t be successful with AI – particularly if you are developing models yourself – unless you understand how these systems actually learn and evolve.

By building a deep understanding of AI technology at the most basic, foundational level, from pre-training to continued pre-training to post-training, you can create or tailor models that are not only intelligent but useful, predictable and optimized for business outcomes.

Successful AI transformation demands an organizational structure that fosters collaboration and minimizes fragmentation. As a centralized AI and Data Division, we view our AI teams as force multipliers, breaking down silos across the organizations. This includes deploying "forward-deployed engineers" – AI experts who integrate directly into business units.

This collaborative model accelerates feedback loops and enables efficient development of tailored tools.

Effective AI transformation hinges on a dynamic feedback loop. For Rakuten, this means creating a "data flywheel" in which humans and AI work synergistically.

By continuously feeding AI models with user input and operational data from trillions of interactions across over 70 businesses and services and then validating and refining AI outputs with human domain experts, we can build unique data assets and improve model reliability.

Meanwhile, we work to digitalize the expertise and know-how of successful customer-facing experts to add to our AI body of knowledge. This integrated approach transforms raw data into a powerful intelligence engine, driving continuous improvement across operations.

AI-nization demands a learning culture. As AI continues to rapidly advance, continuous learning and adaptation are essential.

At Rakuten, this is institutionalized through mechanisms such as our monthly all-team "AI-nization Asakai" meetings that foster alignment, cascade leadership directives and drive engagement.

This commitment to learning from data, experts and experiments helps integrate AI into daily operations, cultivating a dynamic and adaptable organization.

Ultimately, the true measure of AI's success isn't just in its technical sophistication or the volume of its output but in its ability to amplify human potential.

Our journey with AI-nization has shown us that the future success of organizations such as ours is contingent not just on innovating with AI but integrating it for humanity – fostering environments where technology serves to elevate creativity, productivity and connection.

By embracing the “Eight Drivers of AI-nization,” organizations can move beyond the hype, transform their operations and contribute to a future where AI is not merely a tool but a powerful catalyst for a more intelligent, human-centric and thriving society.