Goran Paun

expert panel

For all the talk about AI making customer service faster and more personal, there is still a harder question to answer: Does the customer actually feel better served? An instant answer or perfectly timed recommendation may be useful, but usefulness is not the same as feeling understood. When a customer has a complicated problem, is frustrated or simply wants to talk to someone, efficiency can quickly become the problem.That tension is becoming more important as companies bring AI into more customer interactions. Customers are increasingly comfortable with AI when it saves time or effort, but many still expect a human option when the stakes are higher. The challenge then is figuring out how technology can remove friction without taking away the judgment, reassurance and connection that make good service feel like good service.In this article, members of the Senior Executive AI Think Tank—a curated community of leaders specializing in machine learning, generative AI and enterprise AI applications—explore that challenge from different angles. They discuss personalization and customer memory, where human involvement matters most, how AI can give employees more room for empathy, and how customer signals can shape better products. They also examine what it really means for AI to remember—and how leaders can tell whether an interaction leaves customers feeling understood, rather than simply processed.

expert panel

For executives leading AI transformation, one of the most important decisions is also one of the most difficult: deciding where sensitive data should be processed, stored and governed as artificial intelligence becomes part of the enterprise operating model. The architecture choices organizations make today will shape not only their ability to innovate, but also their ability to protect critical information, meet regulatory expectations and maintain trust.The challenge is that there is no single blueprint for secure AI adoption. Leaders must weigh competing priorities, including the speed and scalability of cloud platforms against the control and data sovereignty of private or hybrid environments, the need for strong governance against the risk of slowing innovation and the benefits of advanced AI capabilities against the responsibility to maintain visibility over how data is used. As AI systems create new layers of information through prompts, outputs, embeddings and logs, organizations must consider not only where data resides, but where it flows and whether they can control its entire lifecycle.Members of the Senior Executive AI Think Tank, a curated group of experts specializing in machine learning, generative AI and enterprise AI applications, examine the most important trade-offs leaders should consider when designing AI architectures for sensitive or regulated data. They also identify common mistakes organizations are making, from focusing only on storage location to overlooking data derivatives, governance gaps and the operational capabilities required to manage AI responsibly. Because architecture decisions are no longer just technical choices—they are business decisions tied to risk, resilience and long-term value.

expert panel

More than 200 economists, AI researchers and technology leaders recently signed the "We Must Act Now" statement, warning that artificial intelligence could reshape the global economy faster than the Industrial Revolution. While the statement calls for stronger institutions and guardrails, it also raises a more immediate question for executives: What should leaders actually do today?Members of the Senior Executive AI Think Tank believe the answer extends well beyond adopting the latest AI tools. Drawing on decades of experience leading enterprise AI, cloud infrastructure, digital transformation and technology strategy, these experts argue that organizations must rethink work itself—redesigning processes, investing in people and establishing governance before AI scales across the enterprise.Their advice comes as organizations accelerate adoption. According to McKinsey's latest State of AI research, AI adoption continues to expand rapidly across industries, with organizations increasingly reporting measurable business impact. Yet the same research also highlights persistent challenges around governance, workforce readiness and risk management, reinforcing the need for thoughtful leadership rather than technology-first decision-making.The following insights from members of the Senior Executive AI Think Tank reveal a consistent message: Organizations that treat AI as a catalyst for better decision-making, stronger employees and more resilient businesses—not simply a cost-cutting exercise—will be best positioned to thrive as AI transforms the economy.

expert panel

Open-weight large language models have become one of enterprise AI's hottest topics. Their promise is compelling: greater control, improved privacy, customization and freedom from vendor lock-in. Yet many executives remain focused on one number—the licensing cost—while overlooking the far larger operational commitment required after deployment.Members of the Senior Executive AI Think Tank, a community of executives and practitioners leading AI strategy across industries, agree that organizations should evaluate open-weight models the same way they would any other mission-critical technology platform: through total cost of ownership, long-term operational resilience and measurable business outcomes.McKinsey has emphasized that successful AI adoption depends not only on selecting the right models but also on building the data infrastructure, operating models and organizational capabilities needed to scale them. For enterprises adopting open-weight models, those ongoing investments can quickly become a larger cost factor than the model itself.In this article, AI Think Tank members examine the hidden costs and strategic considerations behind open-weight AI adoption, sharing how executives should think about infrastructure, talent, security, governance, maintenance and the business cases where owning and operating these models can create a meaningful advantage—and when the hidden costs outweigh the benefits.

expert panel

Artificial intelligence has become remarkably good at creating competent work. It can draft marketing copy, generate product descriptions, design visual assets and even emulate established brand voices in seconds. Yet as organizations adopt many of the same foundation models and workflows, a different challenge is emerging: sameness.Instead of creating stronger differentiation, AI often produces outputs that reflect statistical averages rather than distinctive thinking. The result is an increasing number of websites, advertisements and product messages that feel interchangeable.Members of the Senior Executive AI Think Tank, an invitation-only community of leaders advancing enterprise AI, argue that the real opportunity for differentiation lies far beyond selecting the latest LLM. Across industries ranging from design and marketing to cloud infrastructure and retail technology, they point to a common set of competitive advantages: proprietary knowledge, human judgment, organizational context and leadership that gives AI clear direction.Their insights reveal a fundamental shift in how executives should think about AI strategy. Rather than asking which model is best, organizations should ask what unique expertise, customer understanding and decision-making processes they can bring to those models. The following perspectives explore where lasting competitive advantage is emerging—and why the companies that stand out in the AI era may be the ones that invest most heavily in the capabilities machines can't replicate.

expert panel

When OpenAI unveiled Jalapeño, its first custom AI inference chip developed with Broadcom, the announcement represented more than a hardware milestone. It highlighted a broader shift in the AI industry: the race to make intelligence faster, more affordable and more accessible at scale. As the cost of running large language models declines, product leaders face a new question—not simply what AI can do, but what products become possible when intelligence is inexpensive enough to operate continuously.For much of the generative AI era, product teams have designed around scarcity. They have limited model usage, shortened context windows, reduced reasoning steps and carefully managed AI interactions because every inference call carries a cost. But as custom silicon and AI infrastructure improvements drive down those constraints, AI can move from an occasional feature users activate to an always-present capability embedded throughout workflows. Research from McKinsey & Company estimates that generative AI could create trillions of dollars in annual economic value, but capturing that opportunity will require organizations to integrate AI into core business processes rather than treat it as a standalone tool.Members of the Senior Executive AI Think Tank believe the next generation of AI products will not simply be faster versions of today’s copilots. Below, they explore how OpenAI’s Jalapeño chip could reshape product design, unlock previously uneconomical AI applications and redefine the competitive landscape for organizations building the next generation of intelligent products.