From customer to value

The Battle for Trust: Winning the “Integrated Customer” by Overcoming the GenAI Divide

Q: What does “GenAI Divide” mean?

The **GenAI Divide** occurs when organizations invest heavily in Generative AI initiatives but fail to see a return on their investment. The problem isn’t due to model quality or regulations, but to an outdated strategic approach to implementation. Most GenAI systems aren’t designed to learn from user interaction in a way that meets expectations. As a result, they fail to retain **feedback**, adapt to the specific context of the organization, or improve over time. This failure creates a **“learning gap”** that prevents integration into critical workflows and negates the transformative potential promised by these technologies. To overcome this divide, it’s essential to reform traditional thinking and adopt a more strategic, nuanced approach.

By Enrico Giubertoni

Trusted by

Worldline
Alba Leasing
L'Erbolario
3M

The “Integrated Customer 2025” isn’t just a consumer; they are a “Digital Citizen” with precise, non-negotiable expectations. This emerging behavioral model expects brands to listen, be receptive, and, most importantly, anticipate their desires using Generative AI.

Their expectation is a holistic, seamless experience that integrates online and offline worlds, guaranteeing zero queues, targeted assistance, and needs-focused consultation. For this customer, transparency and clarity aren’t optional—they are fundamental requirements.

The corporate response to these new demands is crucial. Organizations must provide meaningful answers to their target’s present and future needs, creating systems that identify them and anticipate the pace of change. This requires a strategic alignment among top management, internal teams, and external stakeholders.

The GenAI Paradox: The Divide Between Promise and Performance

According to MIT NANDA’s report The GenAI Divide: State of AI in Business 2025 (July 2025), despite an estimated $30-40 billion invested in generative AI initiatives, a staggering 95% of organizations are still failing to achieve adequate returns on their projects. This phenomenon is known as the “GenAI Divide”.

The Antidote to the GenAI Divide: The 4-Way Framework for Victory

To win this battle, organizations must move past the old “build or buy” dichotomy and adopt a more sophisticated approach, known as the Build, Buy, Blend, and Partner Framework, as devised by Deborah Perry Piscione.

1. Build: In-house Autonomy vs. Risk of Failure

The “build” strategy is appropriate when an AI capability represents a core competitive advantage. The example of JPMorgan Chase, which developed a proprietary fraud detection platform, shows how unmatched results can be achieved with in-house solutions. However, this path carries a significantly higher risk of failure than other approaches.

2. Buy: Speed and Expertise vs. Standardization

The “buy” strategy is ideal when time-to-market is crucial and in-house development costs outweigh long-term value. This approach is perfect for standardized functions where competitive advantage comes from implementation excellence rather than unique technology.

3. Blend: Flexibility vs. Integration Complexity

The “blend” approach combines both building and buying. It’s effective when some components require deep customization, while others can be standardized. This hybrid strategy balances speed, cost, and differentiation.

4. Partner: Shared Innovation vs. Third-Party Dependency

Strategic partnerships offer a fourth path, providing comprehensive solutions that include technology, expertise, and ongoing service. Partnerships with customizable, learning-capable tools have a deployment success rate of 67%, significantly outperforming internally built solutions (33%), according to the same MIT NANDA report.

The Only Strategic Imperative: Act Now to Win

To overcome the GenAI Divide, leaders must adopt a sophisticated approach. Successful organizations prioritize buying tools that offer customization and learn over time, empower frontline managers to identify use cases, and focus on automating back-office processes. The key is to demand systems that are capable of learning, deeply integrate into existing workflows, and adapt to feedback.

The window for establishing dominant positions in this evolving AI economy is rapidly closing.

What does “GenAI Divide” mean?

The GenAI Divide occurs when organizations invest heavily in Generative AI initiatives but fail to see a return on their investment. The problem isn’t due to model quality or regulations, but to an outdated strategic approach to implementation.

Most GenAI systems aren’t designed to learn from user interaction in a way that meets expectations. As a result, they fail to retain feedback, adapt to the specific context of the organization, or improve over time. This failure creates a “learning gap” that prevents integration into critical workflows and negates the transformative potential promised by these technologies. To overcome this divide, it’s essential to reform traditional thinking and adopt a more strategic, nuanced approach.