Redis Launches Iris Context Platform as Enterprise AI Moves Beyond RAG Architecture

Quick Facts

• Redis launched Iris context platform Monday to replace traditional RAG architectures for enterprise agentic AI systems
• Enterprise buyer intent for hybrid retrieval systems tripled from 10.3% to 33.3% between January and March 2026
• Traditional RAG systems deliver only 40% accuracy for complex business queries, insufficient for mission-critical applications

Redis launched Iris Monday, a context and memory platform designed to replace traditional retrieval-augmented generation (RAG) architectures as enterprise AI agents push existing systems beyond their limits.

The platform addresses what Redis calls the ‘context problem’ – a fundamental mismatch between RAG systems built for human-scale queries and agentic AI systems that generate orders of magnitude more data requests.

Market Shift Accelerates

VentureBeat’s Q1 2026 VB Pulse RAG Infrastructure Market Tracker found buyer intent to adopt hybrid retrieval systems tripling from 10.3% to 33.3% between January and March. Custom in-house retrieval stacks rose from 24.1% to 35.6% as enterprises outgrew off-the-shelf options.

Retrieval optimization surpassed evaluation as the top enterprise investment priority for the first time, rising from 19% to 28.9% across the quarter.

‘Companies will have orders of magnitude more agents than human beings,’ Redis CEO Rowan Trollope told VentureBeat. ‘Orders of magnitude more agents than human beings means orders of magnitude more load on back end systems.’

RAG Architecture Limitations

Traditional RAG implementations cost enterprises $75,000 to $200,000 but deliver only 40% accuracy rates for complex business queries. The classic RAG model pushes data into agents before models are called, creating bottlenecks when agentic systems require real-time data access.

Production deployments are flipping that architecture. Agents now pull what they need at runtime through tool calls, treating data layers as live resources rather than pre-loaded payloads.

‘It’s just a flip to let the agent pull the data instead of presupposing and stuffing it into the pipeline,’ Trollope said.

Enterprise Adoption Barriers

Critical adoption barriers remain significant. Average implementation costs reach $890,000, while a 340,000 global AI talent shortage compounds challenges. Data infrastructure inadequacy affects 47% of organizations.

Only 11% of organizations have AI agents in production, according to Deloitte’s Emerging Technology Trends study. RAND Corporation research shows more than 80% of AI projects fail to deploy – twice the failure rate of non-AI IT projects.

Stephanie Walter, Practice Leader for AI Stack at HyperFRAME Research, said winning context layers must make agents ‘faster, cheaper, and safer to run.’

‘Agentic AI will not scale in the enterprise if every agent becomes a new cost center, a new data access risk, and a new governance exception,’ Walter said.

Redis Iris Platform Components

Iris combines real-time data ingestion, a semantic interface that auto-generates MCP tools from business data models, and an agent memory server built on Redis Flex. The rewritten storage engine runs 99% of data on flash at one-tenth the cost of in-memory storage alone.

The platform targets enterprises transitioning to context architecture as agentic AI adoption enters its mainstream phase. Projected Fortune 500 deployment reaches 78% in 2026, up from 67% in 2025.

Read more: Context architecture is replacing RAG as agentic AI pushes enterprise retrieval to its limits