AWS Launches OpenSearch Serverless for AI Agents as Bot Traffic Nears Majority by 2027
Updated
Updated · Vocal · Jun 1
AWS Launches OpenSearch Serverless for AI Agents as Bot Traffic Nears Majority by 2027
5 articles · Updated · Vocal · Jun 1
AWS positioned its new OpenSearch Serverless as infrastructure for agentic workloads, arguing AI agents are creating machine-to-machine traffic patterns that differ sharply from human web use.
Those workloads arrive in bursts, trigger multiple concurrent API and retrieval calls, and then drop to zero, making traditional always-on cloud capacity costly and poorly matched.
AWS’s redesign decouples compute from storage so capacity can scale up quickly and scale down to zero when idle, targeting pay-as-you-go economics for agent traffic.
Cloudflare has said bot traffic’s share of HTTP requests is rising and could exceed human traffic in the first half of 2027, while Microsoft, Databricks and Snowflake are also repositioning infrastructure for AI retrieval and state management.
The broader implication is that cloud and enterprise systems may be rebuilt around machine workloads, shifting competition from model releases toward the architecture that can sustain agents in production.
Are small businesses being priced out of an internet re-engineered for giant AI workloads?
As AI agents become the internet's new majority, who is building the safeguards against a catastrophic, system-wide failure?
With the internet being rebuilt for machines, what digital world will be left for humans to inhabit?
AWS OpenSearch Serverless Powers the Agentic Internet: Preparing for AI Agents to Surpass Human Web Traffic by 2027
Overview
AWS has recently announced a major update to its OpenSearch Serverless service, now redesigned to support the growing needs of AI agents. This next-generation platform is built with an agentic-first approach, ensuring that features essential for AI workloads are available from the start. A key upcoming feature is long-term memory for agents, allowing them to retain and recall information over time, which enables more advanced and context-aware interactions. AWS is also adding built-in evaluation and governance, making it easier for developers to build reliable and sophisticated AI agent applications on a scalable and efficient foundation.