Data Poem, a developer of cutting-edge artificial intelligence models for business orchestration, today announced the launch of POEM365, a large causal AI model designed to transform how large enterprise organizations use data for mission-critical decision-making.

POEM365 is an AI model built on the company’s proprietary causal architecture that allows for 90% forecasting precision by building on more than 250 billion integrated transaction records spanning 15,000 brands and $5 trillion in spending. The company said the model was rigorously tested across varied datasets and demonstrates industry-leading accuracy, exceeding the performance of major tech platforms.

Businesses can use POEM365 to understand the “what,” “why” and “how” of their operations, as well as trends and future directions. The company says that, unlike other large language models and time-series models that primarily describe, POEM365 provides deeper analytical insights by revealing the reasons behind events and offering guidance on appropriate responses.

Traditional analytics identify correlations, but the company says POEM365’s “causal AI” capabilities allow it to identify causal relationships, which can provide businesses with strategic insights. Unlike ordinary AI and large language models, it stretches across the entire enterprise and functions to create a unified intelligence layer.

Poem said it designed POEM365 to orchestrate planning and optimization across every function and three organizational time horizons: strategic long-term, meaning 18 to 36 months, annual and monthly agile cycles.

Founder and Chief Executive Bharath Gaddam said the 365 causal architecture is designed to understand “data at scale” to allow businesses to make business-critical decisions with confidence.

“This represents a monumental leap forward in the field of artificial intelligence, which has mostly relied on surface-level statistical correlations,” Gaddam explained. “This architecture applies transfer learning to understand nonlinear data patterns and relationships, synergies and halo effects with near-real-time agility.”

POEM365 includes multiple industry-specific models trained in sector-unique patterns including specialization for automotive, retail, durables, hospitality, e-commerce, electronics, fashion, consumer packaged foods and quick-service dining.

The platform’s core capabilities include multiple AI agents that provide complex data analysis to numerous business workers through an intuitive conversational language interface called the Causal Poets. These different agents include a planning intelligence agent that specializes in forecasting, an optimization intelligence agent that focuses on finding optimal solutions and a research analyst agent that can conduct deep investigations into business problems.

Businesses can adapt the platform to their own needs by using poem-named solutions including Studio, Verse and Rhyme.

Studio is an enterprise-tailored solution that allows businesses to customize the POEM365 model with business data to generate more precise intelligence based on internal knowledge. Verse permits users to deploy AI agents as specialized workers within their network for planning, revenue, demand or wherever they need attention.

Rhyme provides preconfigured optimization capabilities with core business intelligence for organizations getting started with enterprise-wide planning. It includes research agent insights for automated data analysis and basic predictive recommendations.

Image: SiliconANGLE/Microsoft Designer