Cloud infrastructure was largely designed around human activity, such as searching, browsing, streaming and interacting with websites. The rise of AI agents is creating a different type of demand, characterized by rapid bursts of automated activity involving database queries, document searches and API calls. As enterprises deploy more AI-powered systems, cloud providers are adapting infrastructure to support increasingly complex machine-to-machine workloads.

Adapting To The New Age Of Agentic Traffic

Recognizing the fundamental shift in traffic patterns, Amazon Web Services (AWS) has reimagined a foundational element of its cloud offering. On Thursday, AWS launched its next generation of OpenSearch Serverless. This advanced, fully managed search and vector database is engineered specifically for agentic workloads, scaling instantly when task bursts occur and minimizing costs by scaling down to zero during idle periods.

Follow THE FUTURE on LinkedIn, Facebook, Instagram, X and Telegram

Meeting the Demands Of Machine-Generated Traffic

Industry leaders now understand that infrastructure optimized for human-driven internet is ill-suited for the exponential growth of machine-generated traffic. Cloudflare recently reported that bots accounted for 31% of HTTP traffic over the last six months, with AI crawlers and search assistants driving a significant portion of these requests. As Lai Yi Ohlsen, Senior Product Manager at Cloudflare, noted, “Non-human traffic will exceed human traffic sometime in the first half of 2027.”

AI Agents Move Into Production

Recent announcements across the technology sector indicate that AI agents are moving beyond experimentation and into wider commercial use. At Google I/O, Google introduced tools designed to help users delegate tasks such as research and travel planning to AI systems. Businesses are also deploying internal AI agents to automate workflows, increasing the volume of machine-to-machine interactions across enterprise networks.

Technical Changes To OpenSearch

Tia White said the updated platform separates compute resources from storage, allowing capacity to scale more efficiently as demand changes. According to AWS, the model is intended to help organizations manage unpredictable traffic spikes generated by AI systems while reducing infrastructure costs during idle periods.

Integrations and Industry Implications

At launch, OpenSearch Serverless will integrate natively with AI development platforms such as Vercel and Kiro, enabling developers to deploy robust search and vector backends without the overhead of infrastructure management. This innovation aligns with broader industry trends, as companies such as Databricks, Snowflake, Microsoft, and Cloudflare pivot their services to support AI-driven memory and retrieval for enterprise data. As AI adoption accelerates, the pressure for infrastructures that optimize for machine-generated workloads will only intensify.