Guest: Danielle Cook (LinkedIn)
Company: Akamai
Show Name: An Eye on AI
Topic: Edge Computing

Everyone at KubeCon is talking about AI, but how many organizations are actually running AI workloads in production? That’s the uncomfortable question Danielle Cook, Senior Product Marketing Manager at Akamai and CNCF Ambassador, poses in this candid assessment of the current state of AI adoption in cloud native environments. While artificial intelligence dominates conference agendas and vendor booths, Cook suggests the gap between hype and deployment may be wider than the industry wants to admit.

The AI Conversation That Dominates KubeCon

This year’s KubeCon and CloudNativeCon has been unmistakably AI-focused. Nearly every booth features observability solutions enhanced by machine learning, platform engineering tools powered by AI, and infrastructure optimized for training and inference workloads. Cook acknowledges this reality directly, noting that AI and SRE have emerged as the dominant themes across interviews and panel discussions at the event.

But she also raises a critical question that cuts through the enthusiasm. Is AI dominating the conversation because it represents genuine production adoption, or because it’s simply the hot topic everyone wants to associate themselves with? The distinction matters enormously for enterprises trying to separate signal from noise in their infrastructure planning.

The Production Reality Check

Cook points to conflicting data emerging from industry surveys. Some research suggests AI is rapidly taking over as the most transformative and useful tool in enterprise technology. Other surveys paint a more cautious picture, indicating that actual production deployment lags significantly behind the buzz. This divergence reflects a technology landscape in significant flux, where organizations are still figuring out where AI delivers real value versus where it remains experimental.

For technology leaders, this creates a challenging environment. The pressure to adopt AI is intense, driven by competitive concerns and fear of falling behind. Yet rushing into AI deployments without clear use cases or production readiness can lead to wasted resources and disappointed stakeholders. Cook’s perspective suggests that organizations need to be agile both personally and professionally, adapting to rapid change while maintaining realistic expectations about timelines and outcomes.

Akamai’s Strategic Position in Cloud Native AI

Despite her measured assessment of AI hype, Cook is genuinely excited about how Akamai is positioning itself to support the AI future. The company has evolved significantly from its origins as the pioneer of content delivery networks. Through the acquisition of Linode and strategic investments in cloud infrastructure, Akamai has built a presence in the cloud native ecosystem that combines several unique strengths.

The recently announced Akamai Inference Cloud represents the company’s bet on where AI workloads actually need to run. Rather than focusing solely on model training, which tends to happen in centralized cloud environments, Akamai is emphasizing inference at the edge. This approach pulls together the company’s core capabilities in security, edge infrastructure, and managed Kubernetes through their LKE platform.

Cook explains that Akamai’s engineering team has been developing App Platform, an open source tool that integrates CNCF projects to enable organizations to build their own internal developer platforms. This investment reflects the reality that cloud native infrastructure is where AI workloads must run, even if the path from experimentation to production remains uncertain for many organizations.

Why Edge Inference Matters

The emphasis on edge computing for AI inference addresses a fundamental challenge in AI deployment. While training large language models and other AI systems requires massive centralized compute resources, actually using those models in production often demands low latency and local processing. Whether it’s personalized recommendations, real-time fraud detection, or autonomous systems, many AI applications cannot tolerate the round-trip time to distant data centers.

Akamai’s infrastructure, built over decades to deliver content with minimal latency, positions the company to address this edge inference challenge. By combining security tools that protect AI workloads, cloud infrastructure that can scale, and edge locations that bring computation close to users, Akamai offers a differentiated approach compared to hyperscale cloud providers focused primarily on centralized training infrastructure.

What This Signals for Decision-Makers

Cook’s honest assessment of AI hype combined with her enthusiasm for Akamai’s inference strategy offers several signals for technology leaders. First, skepticism about AI deployment timelines is warranted even as organizations should prepare infrastructure for eventual production workloads. Second, the edge will play an increasingly important role as AI moves from training to inference. Third, organizations need platforms that can integrate the growing ecosystem of CNCF projects rather than building everything from scratch.

The cloud native landscape continues evolving rapidly, with AI representing the latest wave of transformation. But as Cook suggests, success requires balancing agility with realism, recognizing that hype cycles and production reality operate on different timelines.