New product autonomously generates optimized AI tools that retrieve and convert enterprise data into precise context for each agentic task, reducing LLM token consumption while enforcing built-in data governance.
Orlando, FL – March 9, 2026 – K2view, a global leader in data management and AI-ready data solution, today announced AI Context Optimizer™, a new engine designed to control LLM inference costs and make agentic AI economically scalable.
As enterprises move from AI experimentation to production-scale agentic systems, a new requirement becomes critical: enterprise data must not only be AI-ready, it must be optimized at the moment of LLM inference to deliver only the precise context required for each task, minimizing token consumption, controlling LLM spend, and limiting unnecessary data exposure.
Agentic AI introduces a new class of applications that interact dynamically with enterprise systems and data. AI agents perform multi-step reasoning while interacting in real time across multiple systems, APIs, and datasets. Each reasoning step may trigger additional data queries, often retrieving large volumes of raw or semi-structured information. When this data is repeatedly sent to LLMs as context, token consumption and latency rise quickly, increasing operational costs and reducing response reliability as models reason over excessive or irrelevant context.
As a result, token usage is rapidly becoming the new cloud bill for enterprise data leaders.
AI Context Optimizer™ addresses this challenge by operating directly on K2view data products to embed cost control and data governance into how AI agents retrieve and update enterprise data. By automatically generating optimized data access tools, it ensures that only precise, task-relevant context is delivered to the LLM at inference time.
Architecting Cost Control and Governance into Agentic AI
Agentic AI systems repeatedly query enterprise systems to gather the information required to complete tasks. When large volumes of enterprise data are sent to LLMs during these iterative interactions, token consumption and latency rise quickly, increasing operational cost.
AI Context Optimizer™ analyzes how agents access enterprise data and detects repetitive, high-cost patterns. It automatically generates optimized data access tools that AI agents invoke instead of repeatedly issuing inefficient requests. These tools return only the precise context required for each task, minimizing the payload sent to LLMs at inference time and significantly reducing token consumption.
These optimized tools embed governance directly into the data interaction layer. Each AI agent is confined to the data of the specific business entity involved in the interaction and restricted to the scope required for the task. Sensitive data is dynamically masked based on user roles and permissions, and agents are restricted to the actions they are authorized to perform.
The result is lower token consumption, reduced latency, controlled AI cost, and built-in policy enforcement.
“Enterprises are excited about the promise of agentic AI, but they’re increasingly concerned about the economics,” said Ronen Schwartz, CEO of K2view. “Intelligence without economic control is not sustainable. Organizations need a way to scale AI responsibly.”
Schwartz added, “AI Context Optimizer™ brings economic discipline to agentic AI by ensuring agents receive only the precise context needed for each task, while enforcing security and governance at runtime. This enables enterprises to deploy agentic systems with cost predictability and governance.”
AI Context Optimizer™ is designed specifically for large enterprises deploying AI agents across complex data environments. As part of K2view GenAI Data Fusion, it adds embedded intelligence that improves how agents retrieve and consume enterprise data without requiring constant manual tuning.
See AI Context Optimizer™ in Action at Gartner Data & Analytics Summit 2026
K2view will demonstrate AI Context Optimizer™ live at the Gartner Data & Analytics Summit 2026. Enterprise data leaders are invited to visit the K2view booth 116 to see how organizations can reduce AI token costs, improve performance, and establish sustainable economic governance for agentic AI systems.