NVIDIA and Accenture Team Up for AI-Focused Business

In October NVIDIA and Accenture announced the launch of the AI-focused Accenture NVIDIA Business Group, with "30,000 professionals receiving training globally to help clients reinvent processes and scale enterprise AI adoption with AI agents." 

In its Q4 earning report for 2024, Accenture shows "Generative AI new bookings of $1 billion for the quarter and $3 billion for the full year." To capitalize on the AI-related serice demands, Accenture launched Accenture AI Refinery division in July 2024. Accenture AI Refinery is built on NVIDIA AI Foundry, laying the ground for this month's partnership.

Lan Guan, Accenture's chief AI officer, said, "Many businesses have not seen the full benefits of generative AI yet. In fact, only 10% of the companies have fully scaled generative AI," Among their challenges are adoption cost, complexity, data, and talent readiness. 

From Generative to Agentic AI

In the advanced press briefing, Accenture and NVIDIA executives revealed Accenture is looking beyond generative AI, and preparing for Agentic AI. "To me, it's no longer about [humans] prompting the LLMs, but ... AI agents that have self-reflecting capabilities," said Guan. "I believe this is the future we are heading towards -- AIs that not just respond but also learn, improve, and work as part of the team. It will change the way we think about productivity." 

In Accenture AI Refinery's hierarchical infrastructure, AI utility agents are overseen by super agents, with orchestrator agents at the top.

The Accenture NVIDIA AI Business Group is aiming to serve those who need marketing reinvention, supply chain reinvention, software-defined factories, healthcare, and immersive consumer experiences, among others. The newly launched business group is expected to help clients lay the foundation for agentic AI functionality, to advance areas such as process reinvention, AI-powered simulation, and sovereign AI (managing cities and governments using AI). 

To test and prove the efficiency of its own offering, Accenture announced its "marketing function is integrating the AI Refinery platform with autonomous agents to help create and run smarter campaigns faster. This will result in a 25-35% reduction in manual steps, 6% cost savings and is expected to achieve a 25-55% increase in speed to market."

NIMs as the Cornerstones of Accenture AI Offerings

Lan Guan clarified that the 30,000 professionals in the group are part of Accenture's global data and AI team. "They've all been trained with NVIDIA technology. They are fully deployed to this initiative, but they will also be working on other ecosystem technologies to take our AI Refinery capabilities to more organizations," she added.

Justin Boitano, NVIDIA's VP of Enterprise AI, said, "NVIDIA NIMs [NVIDIA Inferencing Microservices] are a core part of this initiative. They deliver two to five times better token efficiency when running on our GPUs. You'll have NIMs that read structured data, as well as unstructured data, and NIMs that read domain-specific LLMs."