From promise to making it real: How AI is transforming business operations

The subject of much speculation, the impact of artificial intelligence on business is now quantifiable: our new research shows it is actively reshaping business operations, with tangible outcomes and a striking 1.7x return on investment for early adopters. Read on to learn how to scale AI for lasting impact.

What’s everyone saying?

• Until now, coverage of artificial intelligence (AI) has focused on companies’ first experiments with the technology, described as high potential but also as a not yet realized promise. Last year, half of business leaders globally were still “unclear on AI’s impact despite high investment.”

• Indeed, while many organizations have begun integrating AI into their business operations (e.g., supply chain and procurement, finance and accounting, people operations, and customer operations), there has been limited data so far on whether this has actually improved operational efficiency and outcomes.

What do we have to say?

• Our new report, AI in action: How Gen AI and agentic AI redefine business operations, puts forth a remarkable figure on AI’s operational impact. 

• This is justifiably boosting confidence in AI’s commercial viability, with 40% of organizations now expecting positive ROI within one to three years and another 35% within three to five years. With good reason: in transitioning from experimental proofs of concept to in-production operational AI systems, organizations have witnessed first-hand the benefits AI agents and multi-agent systems can bring, from higher operational efficiency and lower costs to happier customers and fewer errors.

• Consequently, organizations are ramping up investment. This year, 62% of organizations increased their Gen AI investments for business operations. What’s more: of these 62%, 36% are funding these investments not by reallocating existing resources but by dedicating additional budget. In the current economic context, this is no small feat – and it shows just how much of a priority AI investments are to many organizations.

• Gen AI and agentic AI adoption, in particular, are soaring. In 2025, the proportion of organizations with limited or full-scale Gen AI deployment was multiplied by 1.8x, year-on-year, rising from 20% to 36%. Our upcoming report, The rise of agentic AI: How trust holds the key to highly autonomous AI agents, will analyze this ascent of agentic AI and show how humans and AI agents can collaborate to deliver greater benefits.

• This impressive rise in Gen AI adoption has laid the groundwork for agentic AI implementation: the use of AI agents (including multi-agent systems) has more than doubled in one year, with 21% of organizations now using them in operations. And the trend shows no signs of slowing: agentic AI projects are expected to grow by 48% in 2025.

• Gen AI and agentic AI are making a strong impact on operations, with benefits across the board. AI is reshaping business processes and functions such as supply chain management, finance, people operations, and customer operations, delivering significant efficiencies by embedding intelligence into core workflows. In people operations, for example, Gen AI can automate tasks like resume screening and candidate matching, accelerating hiring cycles and lowering recruitment costs.

Who’s doing it right?

For elderly patients facing age-related diseases, early detection can make all the difference in the world. In association with 15 scientific-technological entities, Capgemini developed intelligent solutions that allow patients to benefit from early detection of age-related neurological, motor, and degenerative diseases. These noninvasive systems support decision-making for patients with heart failure, stroke, Parkinson’s, and other diseases, and allow for the evaluation of patient recovery without complex or uncomfortable medical procedures, which can cause anxiety in fragile elderly patients. The bonus? A decentralized approach, through which hospitals, clinics, and institutions can share knowledge without having to transfer sensitive data – meaning patients can trust that their medical information always stays secure and enjoy more precious moments with their loved ones.

What’s the bottom line?

One thing remains clear: in the future, no single process will be handled by one form of AI only. The key will lie in harnessing the strengths of various forms of AI (such as AI-powered intelligent automation, traditional AI or machine learning, Gen AI, and agentic AI) and orchestrating them. That is why, at Capgemini, we help break down these processes and advise on the specific types of AI for most benefit. This is only one of the six essential steps laid out below for advancing toward AI-driven business operations.

• Build a strong foundation of AI readiness, based on leadership advocacy for AI, highly digital business processes and digital twins, governance backed by careful ethical consideration, and enterprise-wide AI literacy.

• Ensure teams are AI-ready to enable smarter human-AI collaboration. This involves rotating employees’ skills, providing training to integrate AI into daily tasks, building efficiency and skillsets, and kicking off a long-term cultural transformation.

• Redesign and reengineer processes to unlock agentic AI’s full potential. To achieve this, organizations should define a broader scope of implementation and follow a structured process transformation approach.

• Embrace agentic AI to complement existing AI applications. Selecting the right use case for each business process is crucial.

• Maintain a sharp focus on cost containment of AI implementation. This means making strategic investments, understanding pricing models, investing in early PoCs, and, most of all, prioritizing solutions that can continuously integrate more cost-efficient AI services as technology evolves.

• Prepare to scale up AI-powered business processes by focusing on making informed, context-specific decisions about whether to build, buy, or adopt a hybrid approach for AI implementation.

Looking for more?

• At Samsung SDS, teams use Intelligent Document Processing to extract data from unstructured receipts, and Gen AI to handle complex, ambiguous cases. Then, an Automation Agent presents the results to a human expert, captures their decision, and feeds it back into the system, creating a continuous learning loop. This human-AI collaboration has led to an 80% reduction in processing time, improved compliance, fewer errors, and more satisfied finance teams!

• In Australia, telco and tech leader Telstra used Microsoft Azure OpenAI Service to empower its frontline staff by developing two Gen AI tools, “One Sentence Summary” and “Ask Telstra.” The result? Employees report significant time savings, improved customer interactions, and greater confidence when handling increasingly complex service requests.

• Swedish appliance manufacturer Electrolux Group leverages Gen AI to enhance its HR and recruitment processes. Automating interview scheduling saves up to 78% time in interview coordination, and the company has registered an 84% increase in application conversion rate.

• From improving citizen services to supporting criminal justice, healthcare, and education or relieving public servants from the "routine stuff", David Knott, Chief Technology Officer of the UK Government, walks us through the impact AI could have (and is already having) on government operations in the latest installment of Conversations for tomorrow.

And you, what are you saying?

Is your organization leveraging Gen AI or agentic AI to transform business operations? Have you witnessed the significant ROI that early adopters are observing? Share your experiences, success stories, and questions in the comments below!